Relevant TOCs

IEEE Transactions on Image Processing - new TOC (2018 February 15) [Website]

Wujie Zhou;Lu Yu;Yang Zhou;Weiwei Qiu;Ming-Wei Wu;Ting Luo; "Local and Global Feature Learning for Blind Quality Evaluation of Screen Content and Natural Scene Images," vol.27(5), pp.2086-2095, May 2018. The blind quality evaluation of screen content images (SCIs) and natural scene images (NSIs) has become an important, yet very challenging issue. In this paper, we present an effective blind quality evaluation technique for SCIs and NSIs based on a dictionary of learned local and global quality features. First, a local dictionary is constructed using local normalized image patches and conventional <inline-formula> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula>-means clustering. With this local dictionary, the learned local quality features can be obtained using a locality-constrained linear coding with max pooling. To extract the learned global quality features, the histogram representations of binary patterns are concatenated to form a global dictionary. The collaborative representation algorithm is used to efficiently code the learned global quality features of the distorted images using this dictionary. Finally, kernel-based support vector regression is used to integrate these features into an overall quality score. Extensive experiments involving the proposed evaluation technique demonstrate that in comparison with most relevant metrics, the proposed blind metric yields significantly higher consistency in line with subjective fidelity ratings.

Jiajian Zeng;Siyuan Liu;Xi Li;Debbah Abderrahmane Mahdi;Fei Wu;Gang Wang; "Deep Context-Sensitive Facial Landmark Detection With Tree-Structured Modeling," vol.27(5), pp.2096-2107, May 2018. Facial landmark detection is typically cast as a point-wise regression problem that focuses on how to build an effective image-to-point mapping function. In this paper, we propose an end-to-end deep learning approach for contextually discriminative feature construction together with effective facial structure modeling. The proposed learning approach is able to predict more contextually discriminative facial landmarks by capturing their associated contextual information. Moreover, we present a tree model to characterize human face structure and a structural loss function to measure the deformation cost between the ground-truth and predicted tree model, which are further incorporated into the proposed learning approach and jointly optimized within a unified framework. The presented tree model is able to well characterize the spatial layout patterns of facial landmarks for capturing the facial structure information. Experimental results demonstrate the effectiveness of the proposed approach against the state-of-the-art over the MTFL and AFLW-full data sets.

Li He;Hong Zhang; "Kernel K-Means Sampling for Nyström Approximation," vol.27(5), pp.2108-2120, May 2018. A fundamental problem in Nyström-based kernel matrix approximation is the sampling method by which training set is built. In this paper, we suggest to use kernel <inline-formula> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula>-means sampling, which is shown in our works to minimize the upper bound of a matrix approximation error. We first propose a unified kernel matrix approximation framework, which is able to describe most existing Nyström approximations under many popular kernels, including Gaussian kernel and polynomial kernel. We then show that, the matrix approximation error upper bound, in terms of the Frobenius norm, is equal to the <inline-formula> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula>-means error of data points in kernel space plus a constant. Thus, the <inline-formula> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula>-means centers of data in kernel space, or the kernel <inline-formula> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula>-means centers, are the optimal representative points with respect to the Frobenius norm error upper bound. Experimental results, with both Gaussian kernel and polynomial kernel, on real-world data sets and image segmentation tasks show the superiority of the proposed method over the state-of-the-art methods.

He Zhang;Vishal M. Patel; "Convolutional Sparse and Low-Rank Coding-Based Image Decomposition," vol.27(5), pp.2121-2133, May 2018. We propose novel convolutional sparse and low-rank coding-based methods for cartoon and texture decomposition. In our method, we first learn a set of generic filters that can efficiently represent cartoon-and texture-type images. Then, using these learned filters, we propose two optimization frameworks to decompose a given image into cartoon and texture components: convolutional sparse coding-based image decomposition; and convolutional low-rank coding-based image decomposition. By working directly on the whole image, the proposed image separation algorithms do not need to divide the image into overlapping patches for leaning local dictionaries. The shift-invariance property is directly modeled into the objective function for learning filters. Extensive experiments show that the proposed methods perform favorably compared with state-of-the-art image separation methods.

Jingjing Meng;Suchen Wang;Hongxing Wang;Junsong Yuan;Yap-Peng Tan; "Video Summarization Via Multiview Representative Selection," vol.27(5), pp.2134-2145, May 2018. Video contents are inherently heterogeneous. To exploit different feature modalities in a diverse video collection for video summarization, we propose to formulate the task as a multiview representative selection problem. The goal is to select visual elements that are representative of a video consistently across different views (i.e., feature modalities). We present in this paper the multiview sparse dictionary selection with centroid co-regularization method, which optimizes the representative selection in each view, and enforces that the view-specific selections to be similar by regularizing them towards a consensus selection. We also introduce a diversity regularizer to favor a selection of diverse representatives. The problem can be efficiently solved by an alternating minimizing optimization with the fast iterative shrinkage thresholding algorithm. Experiments on synthetic data and benchmark video datasets validate the effectiveness of the proposed approach for video summarization, in comparison with other video summarization methods and representative selection methods such as K-medoids, sparse dictionary selection, and multiview clustering.

M. Shahzeb Khan Gul;Bahadir K. Gunturk; "Spatial and Angular Resolution Enhancement of Light Fields Using Convolutional Neural Networks," vol.27(5), pp.2146-2159, May 2018. Light field imaging extends the traditional photography by capturing both spatial and angular distribution of light, which enables new capabilities, including post-capture refocusing, post-capture aperture control, and depth estimation from a single shot. Micro-lens array (MLA) based light field cameras offer a cost-effective approach to capture light field. A major drawback of MLA based light field cameras is low spatial resolution, which is due to the fact that a single image sensor is shared to capture both spatial and angular information. In this paper, we present a learning based light field enhancement approach. Both spatial and angular resolution of captured light field is enhanced using convolutional neural networks. The proposed method is tested with real light field data captured with a Lytro light field camera, clearly demonstrating spatial and angular resolution improvement.

Bin Gao;Xiaoqing Li;Wai Lok Woo;Gui yun Tian; "Physics-Based Image Segmentation Using First Order Statistical Properties and Genetic Algorithm for Inductive Thermography Imaging," vol.27(5), pp.2160-2175, May 2018. Thermographic inspection has been widely applied to non-destructive testing and evaluation with the capabilities of rapid, contactless, and large surface area detection. Image segmentation is considered essential for identifying and sizing defects. To attain a high-level performance, specific physics-based models that describe defects generation and enable the precise extraction of target region are of crucial importance. In this paper, an effective genetic first-order statistical image segmentation algorithm is proposed for quantitative crack detection. The proposed method automatically extracts valuable spatial-temporal patterns from unsupervised feature extraction algorithm and avoids a range of issues associated with human intervention in laborious manual selection of specific thermal video frames for processing. An internal genetic functionality is built into the proposed algorithm to automatically control the segmentation threshold to render enhanced accuracy in sizing the cracks. Eddy current pulsed thermography will be implemented as a platform to demonstrate surface crack detection. Experimental tests and comparisons have been conducted to verify the efficacy of the proposed method. In addition, a global quantitative assessment index F-score has been adopted to objectively evaluate the performance of different segmentation algorithms.

Yigong Zhang;Yingna Su;Jian Yang;Jean Ponce;Hui Kong; "When Dijkstra Meets Vanishing Point: A Stereo Vision Approach for Road Detection," vol.27(5), pp.2176-2188, May 2018. In this paper, we propose a vanishing-point constrained Dijkstra road model for road detection in a stereo-vision paradigm. First, the stereo-camera is used to generate the u- and v-disparity maps of road image, from which the horizon can be extracted. With the horizon and ground region constraints, we can robustly locate the vanishing point of road region. Second, a weighted graph is constructed using all pixels of the image, and the detected vanishing point is treated as the source node of the graph. By computing a vanishing-point constrained Dijkstra minimum-cost map, where both disparity and gradient of gray image are used to calculate cost between two neighbor pixels, the problem of detecting road borders in image is transformed into that of finding two shortest paths that originate from the vanishing point to two pixels in the last row of image. The proposed approach has been implemented and tested over 2600 grayscale images of different road scenes in the KITTI data set. The experimental results demonstrate that this training-free approach can detect horizon, vanishing point, and road regions very accurately and robustly. It can achieve promising performance.

Monjoy Saha;Chandan Chakraborty; "Her2Net: A Deep Framework for Semantic Segmentation and Classification of Cell Membranes and Nuclei in Breast Cancer Evaluation," vol.27(5), pp.2189-2200, May 2018. We present an efficient deep learning framework for identifying, segmenting, and classifying cell membranes and nuclei from human epidermal growth factor receptor-2 (HER2)-stained breast cancer images with minimal user intervention. This is a long-standing issue for pathologists because the manual quantification of HER2 is error-prone, costly, and time-consuming. Hence, we propose a deep learning-based HER2 deep neural network (Her2Net) to solve this issue. The convolutional and deconvolutional parts of the proposed Her2Net framework consisted mainly of multiple convolution layers, max-pooling layers, spatial pyramid pooling layers, deconvolution layers, up-sampling layers, and trapezoidal long short-term memory (TLSTM). A fully connected layer and a softmax layer were also used for classification and error estimation. Finally, HER2 scores were calculated based on the classification results. The main contribution of our proposed Her2Net framework includes the implementation of TLSTM and a deep learning framework for cell membrane and nucleus detection, segmentation, and classification and HER2 scoring. Our proposed Her2Net achieved 96.64% precision, 96.79% recall, 96.71% F-score, 93.08% negative predictive value, 98.33% accuracy, and a 6.84% false-positive rate. Our results demonstrate the high accuracy and wide applicability of the proposed Her2Net in the context of HER2 scoring for breast cancer evaluation.

Ling-Yu Duan;Wei Sun;Xinfeng Zhang;Shiqi Wang;Jie Chen;Jianxiong Yin;Simon See;Tiejun Huang;Alex C. Kot;Wen Gao; "Fast MPEG-CDVS Encoder With GPU-CPU Hybrid Computing," vol.27(5), pp.2201-2216, May 2018. The compact descriptors for visual search (CDVS) standard from ISO/IEC moving pictures experts group has succeeded in enabling the interoperability for efficient and effective image retrieval by standardizing the bitstream syntax of compact feature descriptors. However, the intensive computation of a CDVS encoder unfortunately hinders its widely deployment in industry for large-scale visual search. In this paper, we revisit the merits of low complexity design of CDVS core techniques and present a very fast CDVS encoder by leveraging the massive parallel execution resources of graphics processing unit (GPU). We elegantly shift the computation-intensive and parallel-friendly modules to the state-of-the-arts GPU platforms, in which the thread block allocation as well as the memory access mechanism are jointly optimized to eliminate performance loss. In addition, those operations with heavy data dependence are allocated to CPU for resolving the extra but non-necessary computation burden for GPU. Furthermore, we have demonstrated the proposed fast CDVS encoder can work well with those convolution neural network approaches which enables to leverage the advantages of GPU platforms harmoniously, and yield significant performance improvements. Comprehensive experimental results over benchmarks are evaluated, which has shown that the fast CDVS encoder using GPU-CPU hybrid computing is promising for scalable visual search.

Oded Green; "Efficient Scalable Median Filtering Using Histogram-Based Operations," vol.27(5), pp.2217-2228, May 2018. Median filtering is a smoothing technique for noise removal in images. While there are various implementations of median filtering for a single-core CPU, there are few implementations for accelerators and multi-core systems. Many parallel implementations of median filtering use a sorting algorithm for rearranging the values within a filtering window and taking the median of the sorted value. While using sorting algorithms allows for simple parallel implementations, the cost of the sorting becomes prohibitive as the filtering windows grow. This makes such algorithms, sequential and parallel alike, inefficient. In this work, we introduce the first software parallel median filtering that is non-sorting-based. The new algorithm uses efficient histogram-based operations. These reduce the computational requirements of the new algorithm while also accessing the image fewer times. We show an implementation of our algorithm for both the CPU and NVIDIA’s CUDA supported graphics processing unit (GPU). The new algorithm is compared with several other leading CPU and GPU implementations. The CPU implementation has near perfect linear scaling with a <inline-formula> <tex-math notation="LaTeX">$3.7times $ </tex-math></inline-formula> speedup on a quad-core system. The GPU implementation is several orders of magnitude faster than the other GPU implementations for mid-size median filters. For small kernels, <inline-formula> <tex-math notation="LaTeX">$3 times 3$ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$5 times 5$ </tex-math></inline-formula>, comparison-based approaches are preferable as fewer operations are required. Lastly, the new algorithm is open-source and can be found in the OpenCV library.

Jiachao Zhang;Jie Jia;Andong Sheng;Keigo Hirakawa; "Pixel Binning for High Dynamic Range Color Image Sensor Using Square Sampling Lattice," vol.27(5), pp.2229-2241, May 2018. We propose a new pixel binning scheme for color image sensors. We minimized distortion caused by binning by requiring that the superpixels lie on a square sampling lattice. The proposed binning schemes achieve the equivalent of 4.42 times signal strength improvement with the image resolution loss of 5 times, higher in noise performance and in resolution than the existing binning schemes. As a result, the proposed binning has considerably less artifacts and better noise performance compared with the existing binning schemes. In addition, we provide an extension to the proposed binning scheme for performing single-shot high dynamic range image acquisition.

Yuan Zhou;Anand Rangarajan;Paul D. Gader; "A Gaussian Mixture Model Representation of Endmember Variability in Hyperspectral Unmixing," vol.27(5), pp.2242-2256, May 2018. Hyperspectral unmixing while considering endmember variability is usually performed by the normal compositional model, where the endmembers for each pixel are assumed to be sampled from unimodal Gaussian distributions. However, in real applications, the distribution of a material is often not Gaussian. In this paper, we use Gaussian mixture models (GMM) to represent endmember variability. We show, given the GMM starting premise, that the distribution of the mixed pixel (under the linear mixing model) is also a GMM (and this is shown from two perspectives). The first perspective originates from random variable transformations and gives a conditional density function of the pixels given the abundances and GMM parameters. With proper smoothness and sparsity prior constraints on the abundances, the conditional density function leads to a standard maximum a posteriori (MAP ) problem which can be solved using generalized expectation maximization. The second perspective originates from marginalizing over the endmembers in the GMM, which provides us with a foundation to solve for the endmembers at each pixel. Hence, compared to the other distribution based methods, our model can not only estimate the abundances and distribution parameters, but also the distinct endmember set for each pixel. We tested the proposed GMM on several synthetic and real datasets, and showed its potential by comparing it to current popular methods.

Deepti Ghadiyaram;Janice Pan;Alan C. Bovik; "Learning a Continuous-Time Streaming Video QoE Model," vol.27(5), pp.2257-2271, May 2018. Over-the-top adaptive video streaming services are frequently impacted by fluctuating network conditions that can lead to rebuffering events (stalling events) and sudden bitrate changes. These events visually impact video consumers’ quality of experience (QoE) and can lead to consumer churn. The development of models that can accurately predict viewers’ instantaneous subjective QoE under such volatile network conditions could potentially enable the more efficient design of quality-control protocols for media-driven services, such as YouTube, Amazon, Netflix, and so on. However, most existing models only predict a single overall QoE score on a given video and are based on simple global video features, without accounting for relevant aspects of human perception and behavior. We have created a QoE evaluator, called the time-varying QoE Indexer, that accounts for interactions between stalling events, analyzes the spatial and temporal content of a video, predicts the perceptual video quality, models the state of the client-side data buffer, and consequently predicts continuous-time quality scores that agree quite well with human opinion scores. The new QoE predictor also embeds the impact of relevant human cognitive factors, such as memory and recency, and their complex interactions with the video content being viewed. We evaluated the proposed model on three different video databases and attained standout QoE prediction performance.

Shaofan Lai;Wei-Shi Zheng;Jian-Fang Hu;Jianguo Zhang; "Global-Local Temporal Saliency Action Prediction," vol.27(5), pp.2272-2285, May 2018. Action prediction on a partially observed action sequence is a very challenging task. To address this challenge, we first design a global-local distance model, where a global-temporal distance compares subsequences as a whole and local-temporal distance focuses on individual segment. Our distance model introduces temporal saliency for each segment to adapt its contribution. Finally, a global-local temporal action prediction model is formulated in order to jointly learn and fuse these two types of distances. Such a prediction model is capable of recognizing action of: 1) an on-going sequence and 2) a sequence with arbitrarily frames missing between the beginning and end (known as gap-filling). Our proposed model is tested and compared with related action prediction models on BIT, UCF11, and HMDB data sets. The results demonstrated the effectiveness of our proposal. In particular, we showed the benefit of our proposed model on predicting unseen action types and the advantage on addressing the gapfilling problem as compared with recently developed action prediction models.

Xiatian Zhu;Botong Wu;Dongcheng Huang;Wei-Shi Zheng; "Fast Open-World Person Re-Identification," vol.27(5), pp.2286-2300, May 2018. Existing person re-identification (re-id) methods typically assume that: 1) any probe person is guaranteed to appear in the gallery target population during deployment (i.e., closed-world) and 2) the probe set contains only a limited number of people (i.e., small search scale). Both assumptions are artificial and breached in real-world applications, since the probe population in target people search can be extremely vast in practice due to the ambiguity of probe search space boundary. Therefore, it is unrealistic that any probe person is assumed as one target people, and a large-scale search in person images is inherently demanded. In this paper, we introduce a new person re-id search setting, called large scale open-world (LSOW) re-id, characterized by huge size probe images and open person population in search thus more close to practical deployments. Under LSOW, the under-studied problem of person re-id efficiency is essential in addition to that of commonly studied re-id accuracy. We, therefore, develop a novel fast person re-id method, called Cross-view Identity Correlation and vErification (X-ICE) hashing, for joint learning of cross-view identity representation binarisation and discrimination in a unified manner. Extensive comparative experiments on three large-scale benchmarks have been conducted to validate the superiority and advantages of the proposed X-ICE method over a wide range of the state-of-the-art hashing models, person re-id methods, and their combinations.

Yinzuo Zhou;Luming Zhang;Chao Zhang;Ping Li;Xuelong Li; "Perceptually Aware Image Retargeting for Mobile Devices," vol.27(5), pp.2301-2313, May 2018. Retargeting aims at adapting an original high-resolution photograph/video to a low-resolution screen with an arbitrary aspect ratio. Conventional approaches are generally based on desktop PCs, since the computation might be intolerable for mobile platforms (especially when retargeting videos). Typically, only low-level visual features are exploited, and human visual perception is not well encoded. In this paper, we propose a novel retargeting framework that rapidly shrinks a photograph/video by leveraging human gaze behavior. Specifically, we first derive a geometry-preserving graph ranking algorithm, which efficiently selects a few salient object patches to mimic the human gaze shifting path (GSP) when viewing a scene. Afterward, an aggregation-based CNN is developed to hierarchically learn the deep representation for each GSP. Based on this, a probabilistic model is developed to learn the priors of the training photographs that are marked as aesthetically pleasing by professional photographers. We utilize the learned priors to efficiently shrink the corresponding GSP of a retargeted photograph/video to maximize its similarity to those from the training photographs. Extensive experiments have demonstrated that: 1) our method requires less than 35 ms to retarget a <inline-formula> <tex-math notation="LaTeX">$1024times 768$ </tex-math></inline-formula> photograph (or a <inline-formula> <tex-math notation="LaTeX">$1280times 720$ </tex-math></inline-formula> video frame) on popular iOS/Android devices, which is orders of magnitude faster than the conventional retargeting algorithms; 2) the retargeted photographs/videos produced by our method significantly outperform those of its competitors based on a paired-comparison-based user study; and 3) the learned GSPs are highly indicative of human visual attention according to the human eye tracking experiments.

IEEE Transactions on Medical Imaging - new TOC (2018 February 15) [Website]

* "Table of contents," vol.37(2), pp.C1-C4, Feb. 2018.* Presents the table of contents for this issue of this publication.

* "IEEE Transactions on Medical Imaging publication information," vol.37(2), pp.C2-C2, Feb. 2018.* Provides a listing of current staff, committee members and society officers.

Yang Zhang;Yuexin Guo;Wei-Ning Lee; "Ultrafast Ultrasound Imaging Using Combined Transmissions With Cross-Coherence-Based Reconstruction," vol.37(2), pp.337-348, Feb. 2018. Plane-wave-based ultrafast imaging has become the prevalent technique for non-conventional ultrasound imaging. The image quality, especially in terms of the suppression of artifacts, is generally compromised by reducing the number of transmissions for a higher frame rate. We hereby propose a new ultrafast imaging framework that reduces not only the side lobe artifacts but also the axial lobe artifacts using combined transmissions with a new coherence-based factor. The results from simulations, in vitro wire phantoms, the ex vivo porcine artery, and the in vivo porcine heart show that our proposed methodology greatly reduced the axial lobe artifact by 25±5 dB compared with coherent plane-wave compounding (CPWC), which was considered as the ultrafast imaging standard, and suppressed side lobe artifacts by 15 ± 5 dB compared with CPWC and coherent spherical-wave compounding. The reduction of artifacts in our proposed ultrafast imaging framework led to a better boundary delineation of soft tissues than CPWC.

Hao Gong;Bin Li;Xun Jia;Guohua Cao; "Physics Model-Based Scatter Correction in Multi-Source Interior Computed Tomography," vol.37(2), pp.349-360, Feb. 2018. Multi-source interior computed tomography (CT) has a great potential to provide ultra-fast and organ-oriented imaging at low radiation dose. However, X-ray cross scattering from multiple simultaneously activated X-ray imaging chains compromises imaging quality. Previously, we published two hardware-based scatter correction methods for multi-source interior CT. Here, we propose a software-based scatter correction method, with the benefit of no need for hardware modifications. The new method is based on a physics model and an iterative framework. The physics model was derived analytically, and was used to calculate X-ray scattering signals in both forward direction and cross directions in multi-source interior CT. The physics model was integrated to an iterative scatter correction framework to reduce scatter artifacts. The method was applied to phantom data from both Monte Carlo simulations and physical experimentation that were designed to emulate the image acquisition in a multi-source interior CT architecture recently proposed by our team. The proposed scatter correction method reduced scatter artifacts significantly, even with only one iteration. Within a few iterations, the reconstructed images fast converged toward the “scatter-free” reference images. After applying the scatter correction method, the maximum CT number error at the region-of-interests (ROIs) was reduced to 46 HU in numerical phantom dataset and 48 HU in physical phantom dataset respectively, and the contrast-noise-ratio at those ROIs increased by up to 44.3% and up to 19.7%, respectively. The proposed physics model-based iterative scatter correction method could be useful for scatter correction in dual-source or multi-source CT.

Sungsoo Ha;Klaus Mueller; "A Look-Up Table-Based Ray Integration Framework for 2-D/3-D Forward and Back Projection in X-Ray CT," vol.37(2), pp.361-371, Feb. 2018. Iterative algorithms have become increasingly popular in computed tomography (CT) image reconstruction, since they better deal with the adverse image artifacts arising from low radiation dose image acquisition. But iterative methods remain computationally expensive. The main cost emerges in the projection and back projection operations, where accurate CT system modeling can greatly improve the quality of the reconstructed image. We present a framework that improves upon one particular aspect—the accurate projection of the image basis functions. It differs from current methods in that it substitutes the high computational complexity associated with accurate voxel projection by a small number of memory operations. Coefficients are computed in advance and stored in look-up tables parameterized by the CT system’s projection geometry. The look-up tables only require a few kilobytes of storage and can be efficiently accelerated on the GPU. We demonstrate our framework with both numerical and clinical experiments and compare its performance with the current state-of-the-art scheme—the separable footprint method.

Baudouin Denis de Senneville;Anthony Novell;Chloé Arthuis;Vanda Mendes;Paul-Armand Dujardin;Frédéric Patat;Ayache Bouakaz;Jean-Michel Escoffre;Franck Perrotin; "Development of a Fluid Dynamic Model for Quantitative Contrast-Enhanced Ultrasound Imaging," vol.37(2), pp.372-383, Feb. 2018. Contrast-enhanced ultrasound (CEUS) is a non-invasive imaging technique extensively used for blood perfusion imaging of various organs. This modality is based on the acoustic detection of gas-filled microbubble contrast agents used as intravascular flow tracers. Recent efforts aim at quantifying parameters related to the enhancement in the vascular compartment using time-intensity curve (TIC), and at using these latter as indicators for several pathological conditions. However, this quantification is mainly hampered by two reasons: first, the quantification intrinsically solely relies on temporal intensity variation, the explicit spatial transport of the contrast agent being left out. Second, the exact relationship between the acquired US-signal and the local microbubble concentration is hardly accessible. This paper introduces the use of a fluid dynamic model for the analysis of dynamic CEUS (DCEUS), in order to circumvent the two above-mentioned limitations. A new kinetic analysis is proposed in order to quantify the velocity amplitude of the bolus arrival. The efficiency of proposed methodology is evaluated both in-vitro, for the quantitative estimation of microbubble flow rates, and in-vivo, for the classification of placental insufficiency (control versus ligature) of pregnant rats from DCEUS. Besides, for the in-vivo experimental setup, we demonstrated that the proposed approach outperforms the performance of existing TIC-based methods.

Ozan Oktay;Enzo Ferrante;Konstantinos Kamnitsas;Mattias Heinrich;Wenjia Bai;Jose Caballero;Stuart A. Cook;Antonio de Marvao;Timothy Dawes;Declan P. O‘Regan;Bernhard Kainz;Ben Glocker;Daniel Rueckert; "Anatomically Constrained Neural Networks (ACNNs): Application to Cardiac Image Enhancement and Segmentation," vol.37(2), pp.384-395, Feb. 2018. Incorporation of prior knowledge about organ shape and location is key to improve performance of image analysis approaches. In particular, priors can be useful in cases where images are corrupted and contain artefacts due to limitations in image acquisition. The highly constrained nature of anatomical objects can be well captured with learning-based techniques. However, in most recent and promising techniques such as CNN-based segmentation it is not obvious how to incorporate such prior knowledge. State-of-the-art methods operate as pixel-wise classifiers where the training objectives do not incorporate the structure and inter-dependencies of the output. To overcome this limitation, we propose a generic training strategy that incorporates anatomical prior knowledge into CNNs through a new regularisation model, which is trained end-to-end. The new framework encourages models to follow the global anatomical properties of the underlying anatomy (e.g. shape, label structure) via learnt non-linear representations of the shape. We show that the proposed approach can be easily adapted to different analysis tasks (e.g. image enhancement, segmentation) and improve the prediction accuracy of the state-of-the-art models. The applicability of our approach is shown on multi-modal cardiac data sets and public benchmarks. In addition, we demonstrate how the learnt deep models of 3-D shapes can be interpreted and used as biomarkers for classification of cardiac pathologies.

Amicie de Pierrefeu;Tommy Löfstedt;Fouad Hadj-Selem;Mathieu Dubois;Renaud Jardri;Thomas Fovet;Philippe Ciuciu;Vincent Frouin;Edouard Duchesnay; "Structured Sparse Principal Components Analysis With the TV-Elastic Net Penalty," vol.37(2), pp.396-407, Feb. 2018. Principal component analysis (PCA) is an exploratory tool widely used in data analysis to uncover the dominant patterns of variability within a population. Despite its ability to represent a data set in a low-dimensional space, PCA’s interpretability remains limited. Indeed, the components produced by PCA are often noisy or exhibit no visually meaningful patterns. Furthermore, the fact that the components are usually non-sparse may also impede interpretation, unless arbitrary thresholding is applied. However, in neuroimaging, it is essential to uncover clinically interpretable phenotypic markers that would account for the main variability in the brain images of a population. Recently, some alternatives to the standard PCA approach, such as sparse PCA (SPCA), have been proposed, their aim being to limit the density of the components. Nonetheless, sparsity alone does not entirely solve the interpretability problem in neuroimaging, since it may yield scattered and unstable components. We hypothesized that the incorporation of prior information regarding the structure of the data may lead to improved relevance and interpretability of brain patterns. We therefore present a simple extension of the popular PCA framework that adds structured sparsity penalties on the loading vectors in order to identify the few stable regions in the brain images that capture most of the variability. Such structured sparsity can be obtained by combining, e.g., <inline-formula> <tex-math notation="LaTeX">$ell _{1}$ </tex-math></inline-formula> and total variation (TV) penalties, where the TV regularization encodes information on the underlying structure of the data. This paper presents the structured SPCA (denoted SPCA-TV) optimization framework and its resolution. We demonstrate SPCA-TV’s effectiveness and versatility on three different data sets. It can be applied to any kind of structured data, such as, e.g., <inline-formula> <tex-math notation="LaTe- ">${N}$ </tex-math></inline-formula>-dimensional array images or meshes of cortical surfaces. The gains of SPCA-TV over unstructured approaches (such as SPCA and ElasticNet PCA) or structured approach (such as GraphNet PCA) are significant, since SPCA-TV reveals the variability within a data set in the form of intelligible brain patterns that are easier to interpret and more stable across different samples.

Josip Marjanovic;Markus Weiger;Jonas Reber;David O. Brunner;Benjamin E. Dietrich;Bertram J. Wilm;Romain Froidevaux;Klaas P. Pruessmann; "Multi-Rate Acquisition for Dead Time Reduction in Magnetic Resonance Receivers: Application to Imaging With Zero Echo Time," vol.37(2), pp.408-416, Feb. 2018. For magnetic resonance imaging of tissues with very short transverse relaxation times, radio-frequency excitation must be immediately followed by data acquisition with fast spatial encoding. In zero-echo-time (ZTE) imaging, excitation is performed while the readout gradient is already on, causing data loss due to an initial dead time. One major dead time contribution is the settling time of the filters involved in signal down-conversion. In this paper, a multi-rate acquisition scheme is proposed to minimize dead time due to filtering. Short filters and high output bandwidth are used initially to minimize settling time. With increasing time since the signal onset, longer filters with better frequency selectivity enable stronger signal decimation. In this way, significant dead time reduction is accomplished at only a slight increase in the overall amount of output data. Multi-rate acquisition was implemented with a two-stage filter cascade in a digital receiver based on a field-programmable gate array. In ZTE imaging in a phantom and in vivo, dead time reduction by multi-rate acquisition is shown to improve image quality and expand the feasible bandwidth while increasing the amount of data collected by only a few percent.

Yik-Kiong Hue;Alexander R. Guimaraes;Ouri Cohen;Erez Nevo;Abraham Roth;Jerome L. Ackerman; "Magnetic Resonance Mediated Radiofrequency Ablation," vol.37(2), pp.417-427, Feb. 2018. To introduce magnetic resonance mediated radiofrequency ablation (MR-RFA), in which the MRI scanner uniquely serves both diagnostic and therapeutic roles. In MR-RFA scanner-induced RF heating is channeled to the ablation site via a Larmor frequency RF pickup device and needle system, and controlled via the pulse sequence. MR-RFA was evaluated with simulation of electric and magnetic fields to predict the increase in local specific-absorption-rate (SAR). Temperature-time profiles were measured for different configurations of the device in agar phantoms and ex vivo bovine liver in a 1.5 T scanner. Temperature rise in MR-RFA was imaged using the proton resonance frequency method validated with fiber-optic thermometry. MR-RFA was performed on the livers of two healthy live pigs. Simulations indicated a near tenfold increase in SAR at the RFA needle tip. Temperature-time profiles depended significantly on the physical parameters of the device although both configurations tested yielded temperature increases sufficient for ablation. Resected livers from live ablations exhibited clear thermal lesions. MR-RFA holds potential for integrating RF ablation tumor therapy with MRI scanning. MR-RFA may add value to MRI with the addition of a potentially disposable ablation device, while retaining MRI’s ability to provide real time procedure guidance and measurement of tissue temperature, perfusion, and coagulation.

Roozbeh Shams;Yiming Xiao;François Hébert;Matthew Abramowitz;Rupert Brooks;Hassan Rivaz; "Assessment of Rigid Registration Quality Measures in Ultrasound-Guided Radiotherapy," vol.37(2), pp.428-437, Feb. 2018. Image guidance has become the standard of care for patient positioning in radiotherapy, where image registration is often a critical step to help manage patient motion. However, in practice, verification of registration quality is often adversely affected by difficulty in manual inspection of 3-D images and time constraint, thus affecting the therapeutic outcome. Therefore, we proposed to employ both bootstrapping and the supervised learning methods of linear discriminant analysis and random forest to help robustly assess registration quality in ultrasound-guided radiotherapy. We validated both approaches using phantom and real clinical ultrasound images, and showed that both performed well for the task. While learning-based techniques offer better accuracy and shorter evaluation time, bootstrapping requires no prior training and has a higher sensitivity.

Yitian Zhao;Yalin Zheng;Yonghuai Liu;Yifan Zhao;Lingling Luo;Siyuan Yang;Tong Na;Yongtian Wang;Jiang Liu; "Automatic 2-D/3-D Vessel Enhancement in Multiple Modality Images Using a Weighted Symmetry Filter," vol.37(2), pp.438-450, Feb. 2018. Automated detection of vascular structures is of great importance in understanding the mechanism, diagnosis, and treatment of many vascular pathologies. However, automatic vascular detection continues to be an open issue because of difficulties posed by multiple factors, such as poor contrast, inhomogeneous backgrounds, anatomical variations, and the presence of noise during image acquisition. In this paper, we propose a novel 2-D/3-D symmetry filter to tackle these challenging issues for enhancing vessels from different imaging modalities. The proposed filter not only considers local phase features by using a quadrature filter to distinguish between lines and edges, but also uses the weighted geometric mean of the blurred and shifted responses of the quadrature filter, which allows more tolerance of vessels with irregular appearance. As a result, this filter shows a strong response to the vascular features under typical imaging conditions. Results based on eight publicly available datasets (six 2-D data sets, one 3-D data set, and one 3-D synthetic data set) demonstrate its superior performance to other state-of-the-art methods.

Dong Liu;Anil Kumar Khambampati;Jiangfeng Du; "A Parametric Level Set Method for Electrical Impedance Tomography," vol.37(2), pp.451-460, Feb. 2018. This paper presents an image reconstruction method based on parametric level set (PLS) method using electrical impedance tomography. The conductivity to be reconstructed was assumed to be piecewise constant and the geometry of the anomaly was represented by a shape-based PLS function, which we represent using Gaussian radial basis functions (GRBF). The representation of the PLS function significantly reduces the number of unknowns, and circumvents many difficulties that are associated with traditional level set (TLS) methods, such as regularization, re-initialization and use of signed distance function. PLS reconstruction results shown in this article are some of the first ones using experimental EIT data. The performance of the PLS method was tested with water tank data for two-phase visualization and with simulations which demonstrate the most popular biomedical application of EIT: lung imaging. In addition, robustness studies of the PLS method w.r.t width of the Gaussian function and GRBF centers were performed on simulated lung imaging data. The experimental and simulation results show that PLS method has significant improvement in image quality compared with the TLS reconstruction.

Armin Rund;Christoph Stefan Aigner;Karl Kunisch;Rudolf Stollberger; "Magnetic Resonance RF Pulse Design by Optimal Control With Physical Constraints," vol.37(2), pp.461-472, Feb. 2018. Optimal control approaches have proved useful in designing RF pulses for large tip-angle applications. A typical challenge for optimal control design is the inclusion of constraints resulting from physiological or technical limitations that assure the realizability of the optimized pulses. In this paper, we show how to treat such inequality constraints, in particular, amplitude constraints on the B1 field, the slice-selective gradient, and its slew rate, as well as constraints on the slice profile accuracy. For the latter, a pointwise profile error and additional phase constraints are prescribed. Here, a penalization method is introduced that corresponds to a higher order tracking instead of the common quadratic tracking. The order is driven to infinity in the course of the optimization. We jointly optimize for the RF and slice-selective gradient waveform. The amplitude constraints on these control variables are treated efficiently by semismooth Newton or quasi-Newton methods. The method is flexible, adapting to many optimization goals. As an application, we reduce the power of refocusing pulses, which is important for spin echo-based applications with a short echo spacing. Here, the optimization method is tested in numerical experiments for reducing the pulse power of simultaneous multislice refocusing pulses. The results are validated by phantom and in-vivo experiments.

Tobias Speidel;Jan Paul;Stefan Wundrak;Volker Rasche; "Quasi-Random Single-Point Imaging Using Low-Discrepancy $k$ -Space Sampling," vol.37(2), pp.473-479, Feb. 2018. Magnetic resonance imaging of short relaxation time spin systems has been a widely discussed topic with serious clinical applications and led to the emergence of fast imaging ultra-short echo-time sequences. Nevertheless, these sequences suffer from image blurring, due to the related sampling point spread function and are highly prone to imaging artefacts arising from, e.g., chemical shifts or magnetic susceptibilities. In this paper, we present a concept of spherical quasi-random single-point imaging. The approach is highly accelerateable, due to intrinsic undersampling properties and capable of strong metal artefact suppression. Imaging acceleration is achieved by sampling of quasi-random points in <inline-formula> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula>-space, based on a low-discrepancy sequence, and a combination with non-linear optimization reconstruction techniques [compressed sensing (CS)]. The presented low-discrepancy trajectory shows ideal noise like undersampling properties for the combination with CS, leading to denoised images with excellent metal artefact reduction. Using eightfold undersampling, acquisition time of a few minutes can be achieved for volume acquisitions.

Iñaki Rabanillo;Santiago Aja-Fernández;Carlos Alberola-López;Diego Hernando; "Exact Calculation of Noise Maps and ${g}$ -Factor in GRAPPA Using a ${k}$ -Space Analysis," vol.37(2), pp.480-490, Feb. 2018. Characterization of the noise distribution in magnetic resonance images has multiple applications, including quality assurance and protocol optimization. Noise characterization is particularly important in the presence of parallel imaging acceleration with multi-coil acquisitions, where the noise distribution can contain severe spatial heterogeneities. If the parallel imaging reconstruction is a linear process, an accurate noise analysis can be carried out by taking into account the correlations between all the samples involved. However, for <inline-formula> <tex-math notation="LaTeX">${k}$ </tex-math></inline-formula>-space-based techniques such as generalized autocalibrating partially parallel acquisition (GRAPPA), the exact analysis has been considered computationally prohibitive due to the very large size of the noise covariance matrices required to characterize the noise propagation from <inline-formula> <tex-math notation="LaTeX">${k}$ </tex-math></inline-formula>-space to image space. Previously proposed methods avoid this computational burden by formulating the GRAPPA reconstruction as a pixel-wise linear operation performed in the image space. However, these methods are not exact in the presence of non-uniform sampling of <inline-formula> <tex-math notation="LaTeX">${k}$ </tex-math></inline-formula>-space (e.g., containing a calibration region). For this reason, in this paper, we develop an accurate characterization of the noise distribution for self-calibrated parallel imaging in the presence of arbitrary Cartesian sampling patterns. By exploiting the symmetries and separability in the noise propagation process, the proposed method is computationally efficient and does not require large matrices. Under the assumption of a fixed reconstruction kernel, this method provides the precise distribution of the noise variance for each coil’s image. These coil-by-coil noise maps are subsequently combined according to the coil combination approach- used in image reconstruction, and therefore can be applied with both complex coil combination and root-sum-of-squares approaches. In this paper, we present the proposed noise characterization method and compare it to previous techniques using Monte Carlo simulations as well as phantom acquisitions.

Jo Schlemper;Jose Caballero;Joseph V. Hajnal;Anthony N. Price;Daniel Rueckert; "A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction," vol.37(2), pp.491-503, Feb. 2018. Inspired by recent advances in deep learning, we propose a framework for reconstructing dynamic sequences of 2-D cardiac magnetic resonance (MR) images from undersampled data using a deep cascade of convolutional neural networks (CNNs) to accelerate the data acquisition process. In particular, we address the case where data are acquired using aggressive Cartesian undersampling. First, we show that when each 2-D image frame is reconstructed independently, the proposed method outperforms state-of-the-art 2-D compressed sensing approaches, such as dictionary learning-based MR image reconstruction, in terms of reconstruction error and reconstruction speed. Second, when reconstructing the frames of the sequences jointly, we demonstrate that CNNs can learn spatio-temporal correlations efficiently by combining convolution and data sharing approaches. We show that the proposed method consistently outperforms state-of-the-art methods and is capable of preserving anatomical structure more faithfully up to 11-fold undersampling. Moreover, reconstruction is very fast: each complete dynamic sequence can be reconstructed in less than 10 s and, for the 2-D case, each image frame can be reconstructed in 23 ms, enabling real-time applications.

Chung Chan;John Onofrey;Yiqiang Jian;Mary Germino;Xenophon Papademetris;Richard E. Carson;Chi Liu; "Non-Rigid Event-by-Event Continuous Respiratory Motion Compensated List-Mode Reconstruction for PET," vol.37(2), pp.504-515, Feb. 2018. Respiratory motion during positron emission tomography (PET)/computed tomography (CT) imaging can cause significant image blurring and underestimation of tracer concentration for both static and dynamic studies. In this paper, with the aim to eliminate both intra-cycle and inter-cycle motions, and apply to dynamic imaging, we developed a non-rigid event-by-event (NR-EBE) respiratory motion-compensated list-mode reconstruction algorithm. The proposed method consists of two components: the first component estimates a continuous non-rigid motion field of the internal organs using the internal–external motion correlation. This continuous motion field is then incorporated into the second component, non-rigid MOLAR (NR-MOLAR) reconstruction algorithm to deform the system matrix to the reference location where the attenuation CT is acquired. The point spread function (PSF) and time-of-flight (TOF) kernels in NR-MOLAR are incorporated in the system matrix calculation, and therefore are also deformed according to motion. We first validated NR-MOLAR using a XCAT phantom with a simulated respiratory motion. NR-EBE motion-compensated image reconstruction using both the components was then validated on three human studies injected with 18F-FPDTBZ and one with 18F-fluorodeoxyglucose (FDG) tracers. The human results were compared with conventional non-rigid motion correction using discrete motion field (NR-discrete, one motion field per gate) and a previously proposed rigid EBE motion-compensated image reconstruction (R-EBE) that was designed to correct for rigid motion on a target lesion/organ. The XCAT results demonstrated that NR-MOLAR incorporating both PSF and TOF kernels effectively corrected for non-rigid motion. The 18F-FPDTBZ studies showed that NR-EBE out-performed NR-Discrete, and yielded comparable results with R-EBE on target organs while yielding superior image quality in other regions. The FDG study showed that NR- EBE clearly improved the visibility of multiple moving lesions in the liver where some of them could not be discerned in other reconstructions, in addition to improving quantification. These results show that NR-EBE motion-compensated image reconstruction appears to be a promising tool for lesion detection and quantification when imaging thoracic and abdominal regions using PET.

Enrico Pellegrini;Gavin Robertson;Tom MacGillivray;Jano van Hemert;Graeme Houston;Emanuele Trucco; "A Graph Cut Approach to Artery/Vein Classification in Ultra-Widefield Scanning Laser Ophthalmoscopy," vol.37(2), pp.516-526, Feb. 2018. The classification of blood vessels into arterioles and venules is a fundamental step in the automatic investigation of retinal biomarkers for systemic diseases. In this paper, we present a novel technique for vessel classification on ultra-wide-field-of-view images of the retinal fundus acquired with a scanning laser ophthalmoscope. To the best of our knowledge, this is the first time that a fully automated artery/vein classification technique for this type of retinal imaging with no manual intervention has been presented. The proposed method exploits hand-crafted features based on local vessel intensity and vascular morphology to formulate a graph representation from which a globally optimal separation between the arterial and venular networks is computed by graph cut approach. The technique was tested on three different data sets (one publicly available and two local) and achieved an average classification accuracy of 0.883 in the largest data set.

Andreas Horneff;Michael Eder;Erich Hell;Johannes Ulrici;Jörg Felder;Volker Rasche;Jens Anders; "An EM Simulation-Based Design Flow for Custom-Built MR Coils Incorporating Signal and Noise," vol.37(2), pp.527-535, Feb. 2018. Developing custom-built MR coils is a cumbersome task, in which an a priori prediction of the coils’ SNR performance, their sensitivity pattern, and their depth of penetration helps to greatly speed up the design process by reducing the required hardware manufacturing iterations. The simulation-based design flow presented in this paper takes the entire MR imaging process into account. That is, it includes all geometric and material properties of the coil and the phantom, the thermal noise as well as the target MR sequences. The proposed simulation-driven design flow is validated using a manufactured prototype coil, whose performance was optimized regarding its SNR performance, based on the presented design flow, by comparing the coil’s measured performance against the simulated results. In these experiments, the mean and the standard deviation of the relative error between the simulated and measured coil sensitivity pattern were found to be <inline-formula> <tex-math notation="LaTeX">$mu ={1.79}%$ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$sigma ={3.15}%$ </tex-math></inline-formula>. Moreover, the peak deviation between the simulated and measured voxel SNR was found to be less than 4%, indicating that simulations are in good accordance with the measured results, validating the proposed software-based design approach.

Neerav Dixit;Pascal P. Stang;John M. Pauly;Greig C. Scott; "Thermo-Acoustic Ultrasound for Detection of RF-Induced Device Lead Heating in MRI," vol.37(2), pp.536-546, Feb. 2018. Patients who have implanted medical devices with long conductive leads are often restricted from receiving MRI scans due to the danger of RF-induced heating near the lead tips. Phantom studies have shown that this heating varies significantly on a case-by-case basis, indicating that many patients with implanted devices can receive clinically useful MRI scans without harm. However, the difficulty of predicting RF-induced lead tip heating prior to scanning prevents numerous implant recipients from being scanned. Here, we demonstrate that thermo-acoustic ultrasound (TAUS) has the potential to be utilized for a pre-scan procedure assessing the risk of RF-induced lead tip heating in MRI. A system was developed to detect TAUS signals by four different TAUS acquisition methods. We then integrated this system with an MRI scanner and detected a peak in RF power absorption near the tip of a model lead when transmitting from the scanner’s body coil. We also developed and experimentally validated simulations to characterize the thermo-acoustic signal generated near lead tips. These results indicate that TAUS is a promising method for assessing RF implant safety, and with further development, a TAUS pre-scan could allow many more patients to have access to MRI scans of significant clinical value.

Juan F. P. J. Abascal;Manuel Desco;Juan Parra-Robles; "Incorporation of Prior Knowledge of Signal Behavior Into the Reconstruction to Accelerate the Acquisition of Diffusion MRI Data," vol.37(2), pp.547-556, Feb. 2018. Diffusion MRI data are generally acquired using hyperpolarized gases during patient breath-hold, which yields a compromise between achievable image resolution, lung coverage, and number of <inline-formula> <tex-math notation="LaTeX">$b$ </tex-math></inline-formula>-values. In this paper, we propose a novel method that accelerates the acquisition of diffusion MRI data by undersampling in both the spatial and <inline-formula> <tex-math notation="LaTeX">$b$ </tex-math></inline-formula>-value dimensions and incorporating knowledge about signal decay into the reconstruction (SIDER). SIDER is compared with total variation (TV) reconstruction by assessing its effect on both the recovery of ventilation images and the estimated mean alveolar dimensions (MADs). Both methods are assessed by retrospectively undersampling diffusion data sets (<inline-formula> <tex-math notation="LaTeX">$n $ </tex-math></inline-formula>=8) of healthy volunteers and patients with Chronic Obstructive Pulmonary Disease (COPD) for acceleration factors between x2 and x10. TV led to large errors and artifacts for acceleration factors equal to or larger than x5. SIDER improved TV, with a lower solution error and MAD histograms closer to those obtained from fully sampled data for acceleration factors up to x10. SIDER preserved image quality at all acceleration factors, although images were slightly smoothed and some details were lost at x10. In conclusion, we developed and validated a novel compressed sensing method for lung MRI imaging and achieved high acceleration factors, which can be used to increase the amount of data acquired during breath-hold. This methodology is expected to improve the accuracy of estimated lung microstructure dimensions and provide more options in the study of lung diseases with MRI.

Evan Levine;Brian Hargreaves; "On-the-Fly Adaptive ${k}$ -Space Sampling for Linear MRI Reconstruction Using Moment-Based Spectral Analysis," vol.37(2), pp.557-567, Feb. 2018. In high-dimensional magnetic resonance imaging applications, time-consuming, sequential acquisition of data samples in the spatial frequency domain (<inline-formula> <tex-math notation="LaTeX">${k}$ </tex-math></inline-formula>-space) can often be accelerated by accounting for dependencies in linear reconstruction, at the cost of noise amplification that depends on the sampling pattern. Common examples are support-constrained, parallel, and dynamic MRI, and <inline-formula> <tex-math notation="LaTeX">${k}$ </tex-math></inline-formula>-space sampling strategies are primarily driven by image-domain metrics that are expensive to compute for arbitrary sampling patterns. It remains challenging to provide systematic and computationally efficient automatic designs of arbitrary multidimensional Cartesian sampling patterns that mitigate noise amplification, given the subspace to which the object is confined. To address this problem, this paper introduces a theoretical framework that describes local geometric properties of the sampling pattern and relates them to the spread in the eigenvalues of the information matrix described by its first two spectral moments. This new criterion is then used for very efficient optimization of complex multidimensional sampling patterns that does not require reconstructing images or explicitly mapping noise amplification. Experiments with in vivo data show strong agreement between this criterion and traditional, comprehensive image-domain- and <inline-formula> <tex-math notation="LaTeX">${k}$ </tex-math></inline-formula>-space-based metrics, indicating the potential of the approach for computationally efficient (on-the-fly), automatic, and adaptive design of sampling patterns.

John S. H. Baxter;Zahra Hosseini;Terry M. Peters;Maria Drangova; "Cyclic Continuous Max-Flow: A Third Paradigm in Generating Local Phase Shift Maps in MRI," vol.37(2), pp.568-579, Feb. 2018. Sensitivity to phase deviations in MRI forms the basis of a variety of techniques, including magnetic susceptibility weighted imaging and chemical shift imaging. Current phase processing techniques fall into two families: those which process the complex image data with magnitude and phase coupled, and phase unwrapping-based techniques that first linearize the phase topology across the image. However, issues, such as low signal and the existence of phase poles, can lead both methods to experience error. Cyclic continuous max-flow (CCMF) phase processing uses primal-dual-variational optimization over a cylindrical manifold, which represent the inherent topology of phase images, increasing its robustness to these issues. CCMF represents a third distinct paradigm in phase processing, being the only technique equipped with the inherent topology of phase. CCMF is robust and efficient with at least comparable accuracy as the prior paradigms.

Amar V. Nasrulloh;Chris G. Willcocks;Philip T. G. Jackson;Caspar Geenen;Maged S. Habib;David H. W. Steel;Boguslaw Obara; "Multi-Scale Segmentation and Surface Fitting for Measuring 3-D Macular Holes," vol.37(2), pp.580-589, Feb. 2018. Macular holes are blinding conditions, where a hole develops in the central part of retina, resulting in reduced central vision. The prognosis and treatment options are related to a number of variables, including the macular hole size and shape. High-resolution spectral domain optical coherence tomography allows precise imaging of the macular hole geometry in three dimensions, but the measurement of these by human observers is time-consuming and prone to high inter- and intra-observer variability, being characteristically measured in 2-D rather than 3-D. We introduce several novel techniques to automatically retrieve accurate 3-D measurements of the macular hole, including: surface area, base area, base diameter, top area, top diameter, height, and minimum diameter. Specifically, we introduce a multi-scale 3-D level set segmentation approach based on a state-of-the-art level set method, and we introduce novel curvature-based cutting and 3-D measurement procedures. The algorithm is fully automatic, and we validate our extracted measurements both qualitatively and quantitatively, where our results show the method to be robust across a variety of scenarios. Our automated processes are considered a significant contribution for clinical applications.

Georg Schramm;Martin Holler;Ahmadreza Rezaei;Kathleen Vunckx;Florian Knoll;Kristian Bredies;Fernando Boada;Johan Nuyts; "Evaluation of Parallel Level Sets and Bowsher’s Method as Segmentation-Free Anatomical Priors for Time-of-Flight PET Reconstruction," vol.37(2), pp.590-603, Feb. 2018. In this article, we evaluate Parallel Level Sets (PLS) and Bowsher’s method as segmentation-free anatomical priors for regularized brain positron emission tomography (PET) reconstruction. We derive the proximity operators for two PLS priors and use the EM-TV algorithm in combination with the first order primal-dual algorithm by Chambolle and Pock to solve the non-smooth optimization problem for PET reconstruction with PLS regularization. In addition, we compare the performance of two PLS versions against the symmetric and asymmetric Bowsher priors with quadratic and relative difference penalty function. For this aim, we first evaluate reconstructions of 30 noise realizations of simulated PET data derived from a real hybrid positron emission tomography/magnetic resonance imaging (PET/MR) acquisition in terms of regional bias and noise. Second, we evaluate reconstructions of a real brain PET/MR data set acquired on a GE Signa time-of-flight PET/MR in a similar way. The reconstructions of simulated and real 3D PET/MR data show that all priors were superior to post-smoothed maximum likelihood expectation maximization with ordered subsets (OSEM) in terms of bias-noise characteristics in different regions of interest where the PET uptake follows anatomical boundaries. Our implementation of the asymmetric Bowsher prior showed slightly superior performance compared with the two versions of PLS and the symmetric Bowsher prior. At very high regularization weights, all investigated anatomical priors suffer from the transfer of non-shared gradients.

Valur T. Olafsson;Douglas C. Noll;Jeffrey A. Fessler; "Fast Spatial Resolution Analysis of Quadratic Penalized Least-Squares Image Reconstruction With Separate Real and Imaginary Roughness Penalty: Application to fMRI," vol.37(2), pp.604-614, Feb. 2018. Penalized least-squares iterative image reconstruction algorithms used for spatial resolution-limited imaging, such as functional magnetic resonance imaging (fMRI), commonly use a quadratic roughness penalty to regularize the reconstructed images. When used for complex-valued images, the conventional roughness penalty regularizes the real and imaginary parts equally. However, these imaging methods sometimes benefit from separate penalties for each part. The spatial smoothness from the roughness penalty on the reconstructed image is dictated by the regularization parameter(s). One method to set the parameter to a desired smoothness level is to evaluate the full width at half maximum of the reconstruction method’s local impulse response. Previous work has shown that when using the conventional quadratic roughness penalty, one can approximate the local impulse response using an FFT-based calculation. However, that acceleration method cannot be applied directly for separate real and imaginary regularization. This paper proposes a fast and stable calculation for this case that also uses FFT-based calculations to approximate the local impulse responses of the real and imaginary parts. This approach is demonstrated with a quadratic image reconstruction of fMRI data that uses separate roughness penalties for the real and imaginary parts.

Nikolas Lessmann;Bram van Ginneken;Majd Zreik;Pim A. de Jong;Bob D. de Vos;Max A. Viergever;Ivana Išgum; "Automatic Calcium Scoring in Low-Dose Chest CT Using Deep Neural Networks With Dilated Convolutions," vol.37(2), pp.615-625, Feb. 2018. Heavy smokers undergoing screening with low-dose chest CT are affected by cardiovascular disease as much as by lung cancer. Low-dose chest CT scans acquired in screening enable quantification of atherosclerotic calcifications and thus enable identification of subjects at increased cardiovascular risk. This paper presents a method for automatic detection of coronary artery, thoracic aorta, and cardiac valve calcifications in low-dose chest CT using two consecutive convolutional neural networks. The first network identifies and labels potential calcifications according to their anatomical location and the second network identifies true calcifications among the detected candidates. This method was trained and evaluated on a set of 1744 CT scans from the National Lung Screening Trial. To determine whether any reconstruction or only images reconstructed with soft tissue filters can be used for calcification detection, we evaluated the method on soft and medium/sharp filter reconstructions separately. On soft filter reconstructions, the method achieved F1 scores of 0.89, 0.89, 0.67, and 0.55 for coronary artery, thoracic aorta, aortic valve, and mitral valve calcifications, respectively. On sharp filter reconstructions, the F1 scores were 0.84, 0.81, 0.64, and 0.66, respectively. Linearly weighted kappa coefficients for risk category assignment based on per subject coronary artery calcium were 0.91 and 0.90 for soft and sharp filter reconstructions, respectively. These results demonstrate that the presented method enables reliable automatic cardiovascular risk assessment in all low-dose chest CT scans acquired for lung cancer screening.

Markus Zimmermann;Zaheer Abbas;Krzysztof Dzieciol;N. Jon Shah; "Accelerated Parameter Mapping of Multiple-Echo Gradient-Echo Data Using Model-Based Iterative Reconstruction," vol.37(2), pp.626-637, Feb. 2018. A new reconstruction method, coined MIRAGE, is presented for accurate, fast, and robust parameter mapping of multiple-echo gradient-echo (MEGE) imaging, the basis sequence of novel quantitative magnetic resonance imaging techniques such as water content and susceptibility mapping. Assuming that the temporal signal can be modeled as a sum of damped complex exponentials, MIRAGE performs model-based reconstruction of undersampled data by minimizing the rank of local Hankel matrices. It further incorporates multi-channel information and spatial prior knowledge. Finally, the parameter maps are estimated using nonlinear regression. Simulations and retrospective undersampling of phantom and in vivo data affirm robustness, e.g., to strong inhomogeneity of the static magnetic field and partial volume effects. MIRAGE is compared with a state-of-the-art compressed sensing method, <inline-formula> <tex-math notation="LaTeX">${{text {L}}_{1}}$ </tex-math></inline-formula>-ESPIRiT. Parameter maps estimated from reconstructed data using MIRAGE are shown to be accurate, with the mean absolute error reduced by up to 50% for in vivo results. The proposed method has the potential to improve the diagnostic utility of quantitative imaging techniques that rely on MEGE data.

Ling Zhang;Le Lu;Ronald M. Summers;Electron Kebebew;Jianhua Yao; "Convolutional Invasion and Expansion Networks for Tumor Growth Prediction," vol.37(2), pp.638-648, Feb. 2018. Tumor growth is associated with cell invasion and mass-effect, which are traditionally formulated by mathematical models, namely reaction-diffusion equations and biomechanics. Such models can be personalized based on clinical measurements to build the predictive models for tumor growth. In this paper, we investigate the possibility of using deep convolutional neural networks to directly represent and learn the cell invasion and mass-effect, and to predict the subsequent involvement regions of a tumor. The invasion network learns the cell invasion from information related to metabolic rate, cell density, and tumor boundary derived from multimodal imaging data. The expansion network models the mass-effect from the growing motion of tumor mass. We also study different architectures that fuse the invasion and expansion networks, in order to exploit the inherent correlations among them. Our network can easily be trained on population data and personalized to a target patient, unlike most previous mathematical modeling methods that fail to incorporate population data. Quantitative experiments on a pancreatic tumor data set show that the proposed method substantially outperforms a state-of-the-art mathematical model-based approach in both accuracy and efficiency, and that the information captured by each of the two subnetworks is complementary.

Ben Cassidy;F. DuBois Bowman;Caroline Rae;Victor Solo; "On the Reliability of Individual Brain Activity Networks," vol.37(2), pp.649-662, Feb. 2018. There is intense interest in fMRI research on whole-brain functional connectivity, and however, two fundamental issues are still unresolved: the impact of spatiotemporal data resolution (spatial parcellation and temporal sampling) and the impact of the network construction method on the reliability of functional brain networks. In particular, the impact of spatiotemporal data resolution on the resulting connectivity findings has not been sufficiently investigated. In fact, a number of studies have already observed that functional networks often give different conclusions across different parcellation scales. If the interpretations from functional networks are inconsistent across spatiotemporal scales, then the whole validity of the functional network paradigm is called into question. This paper investigates the consistency of resting state network structure when using different temporal sampling or spatial parcellation, or different methods for constructing the networks. To pursue this, we develop a novel network comparison framework based on persistent homology from a topological data analysis. We use the new network comparison tools to characterize the spatial and temporal scales under which consistent functional networks can be constructed. The methods are illustrated on Human Connectome Project data, showing that the DISCOH2 network construction method outperforms other approaches at most data spatiotemporal resolutions.

Elias Pavlatos;Hong Chen;Keyton Clayson;Xueliang Pan;Jun Liu; "Imaging Corneal Biomechanical Responses to Ocular Pulse Using High-Frequency Ultrasound," vol.37(2), pp.663-670, Feb. 2018. Imaging corneal biomechanical changes or abnormalities is important for better clinical diagnosis and treatment of corneal diseases. We propose a novel ultrasound-based method, called ocular pulse elastography (OPE), to image corneal deformation during the naturally occurring ocular pulse. Experiments on animal and human donor eyes, as well as synthetic radiofrequency (RF) data, were used to evaluate the efficacy of the OPE method. Using very high-frequency ultrasound (center frequency = 55 MHz), correlation-based speckle tracking yielded an accuracy of less than 10% error for axial tissue displacements of <inline-formula> <tex-math notation="LaTeX">$0.5~mu text{m}$ </tex-math></inline-formula> or above. Satisfactory speckle tracking was achieved for out-of-plane displacements up to <inline-formula> <tex-math notation="LaTeX">$32~mu text{m}$ </tex-math></inline-formula>. Using synthetic RF data with or without a pre-defined uniform strain, the OPE method detected strains down to 0.0001 axially and 0.00025 laterally with an error less than 10%. Experiments in human donor eyes showed excellent repeatability with an intraclass correlation of 0.98. The measurement outcome from OPE was also shown to be highly correlated with that of standard inflation. These results suggest the feasibility of OPE as a potential clinical tool for evaluating corneal biomechanics in vivo.

* "40th International Conference of the IEEE Engineering in Medicine and Biology Society," vol.37(2), pp.671-671, Feb. 2018.* Describes the above-named upcoming conference event. May include topics to be covered or calls for papers.

* "IEEE EMBS Micro and Nanotechnology Conference in Medicine," vol.37(2), pp.672-672, Feb. 2018.* Describes the above-named upcoming conference event. May include topics to be covered or calls for papers.

* "IEEE Transactions on Medical Imaging information for authors," vol.37(2), pp.C3-C3, Feb. 2018.* Provides instructions and guidelines to prospective authors who wish to submit manuscripts.

IET Image Processing - new TOC (2018 February 15) [Website]

Badal Soni;Pradip K. Das;Dalton Meitei Thounaojam; "CMFD: a detailed review of block based and key feature based techniques in image copy-move forgery detection," vol.12(2), pp.167-178, 2 2018. With the advancement of image editing tools in today's world, the manipulation of images like cropping, cloning, resizing, etc., becomes an easy proposition and on the other end, checking or determining whether an image has been manipulated or not, becomes a great challenge. Copy-move forgery in images is the most popular tampering method in which a portion of an image is copied and pasted in some other location of the same image. The detection of copy-move forgery has become a prominent research area. This study presents a detailed review and critical discussions with pros and cons of each of copy-move forgery detection techniques from 2007 to 2017. This study also addresses the variation in databases, issues, challenges, future directions and references in this domain.

Malihe Sabeti;Reza Boostani;Bita Davoodi; "Improved particle swarm optimisation to estimate bone age," vol.12(2), pp.179-187, 2 2018. This paper automatizes the process of bone age maturity assessment by applying three versions of particle swarm optimization (PSO) along with image processing methods to the left hand X-ray images. PSO versions were adopted to enhance the segmentation accuracy. The proposed method was compared to the conventional visual inspection method in terms of three segmentation criteria, classification accuracy, robustness against noise and computational complexity. Herein, PSO, worst behavior-based PSO (WB-PSO) and adaptive inertia weight (AIW-PSO) along with Otsu and an iteratively statistical method were implemented to segment the hand radiographs. A dataset containing left hand-wrist radiographs from 65 referred children was collected. Their results provided 82.49, 83.08, 84.27, 81.76 and 69.04% classification accuracy using the PSO, WB-PSO, AIW-PSO, Otsu and the iteratively statistical methods, respectively. To assess the robustness of the implemented methods, white Gaussian noise with different intensities was added to the images and the results indicated that as the noise level increased the robustness against noise for the PSO variants became more highlighted compared to the Otsu and statistical methods. Due to the convincing results, the AIW-PSO image segmentation system is suggested as an auxiliary diagnostic tool to help specialists for more accurate age bone estimation.

Yang-Yi Chen;Chun-Liang Lin;Yu-Cheng Lin;Changchen Zhao; "Non-invasive detection of alcohol concentration based on photoplethysmogram signals," vol.12(2), pp.188-193, 2 2018. Sobriety test, which commonly uses a breathalyser for estimating blood alcohol content (BAC) from a breath sample, is commonly used to detect drunk driving. The detection often gives rise to sanitary concern and violation of human right. This research proposes a new method that employs non-invasive measurement of photoplethysmography (PPG) signal for detecting BAC of the test subject under the sobriety test. The PPG signal is measured via an LED transmitter and a receiver that illuminates finger and measures the changes in lighting. Since a PPG signal contains information of systolic and diastolic blood pressure, it is possible to be used for the purpose of detecting BAC. The authors have developed a practical alcohol sobriety test system to analyse the status of alcohol intake of the subject. Extensive tests have been conducted to examine feasibility of the proposed system with an identification rate up to 85%.

Weiliang Tao;Yan Liu; "Combined imaging matching method of side scan sonar images with prior position knowledge," vol.12(2), pp.194-199, 2 2018. Side scan sonar (SSS) image matching plays an important role in underwater applications, such as combining images to form wide range relief images, underwater simultaneous location and mapping, and construction of images of underwater terrain and objects. Because there are differences between the imaging mechanisms of SSS images and optical images, current matching algorithms for optical images do not always work well for SSS images. This study proposes a combined image matching approach for SSS images. First, the images are preprocessed to remove some environmental effects. Second, feature points that are stable for affine transformations are extracted from the SSS matching images based on speeded-up robust features with prior position knowledge. Third, geometric correction is made by the random sample consensus. After that, a similarity calculation is performed. The proposed approach speeds up the matching process and improves its accuracy by combining feature points matching and similarity calculation. It reduces the mismatching rate and the computation requirement by estimating the location uncertainty with prior knowledge and reducing the searching regions for image matching. Experiments show that the matching algorithm takes less time and has greater reliability and accuracy than traditional algorithms.

Muhammad Attique Khan;Muhammad Sharif;Muhammad Younus Javed;Tallha Akram;Mussarat Yasmin;Tanzila Saba; "License number plate recognition system using entropy-based features selection approach with SVM," vol.12(2), pp.200-209, 2 2018. License plate recognition (LPR) system plays a vital role in security applications which include road traffic monitoring, street activity monitoring, identification of potential threats, and so on. Numerous methods were adopted for LPR but still, there is enough space for a single standard approach which can be able to deal with all sorts of problems such as light variations, occlusion, and multi-views. The proposed approach is an effort to deal under such conditions by incorporating multiple features extraction and fusion. The proposed architecture is comprised of four primary steps: (i) selection of luminance channel from CIE-Lab colour space, (ii) binary segmentation of selected channel followed by image refinement, (iii) a fusion of Histogram of oriented gradients (HOG) and geometric features followed by a selection of appropriate features using a novel entropy-based method, and (iv) features classification with support vector machine (SVM). To authenticate the results of proposed approach, different performance measures are considered. The selected measures are False positive rate (FPR), False negative rate (FNR), and accuracy which is achieved maximum up to 99.5%. Simulation results reveal that the proposed method performs exceptionally better compared with existing works.

Can Chen;Fei Ding;Dengyin Zhang; "Perceptual hash algorithm-based adaptive GOP selection algorithm for distributed compressive video sensing," vol.12(2), pp.210-217, 2 2018. Distributed compressive video sensing (DCVS) is a novel video coding technique that shifts sophisticated motion estimation and compensation from the encoder to the decoder and is suitable for resource-limited communication, namely wireless video sensor networks (WVSNs). In DCVS, key frames serve as the reference for subsequent non-key frames in a given group of pictures (GOP). However, in fast-motion sequences, e.g. scene-changing sequences, a fixed GOP size can cause inaccuracy in the selection of reference key frames. The difference in the peak signal-to-noise ratio between key frames and non-key frames caused by this inaccuracy appears as flicker in the decoded video, negatively affecting the quality of experience. To address this problem, the authors present a perceptual hash algorithm-based adaptive GOP selection algorithm for DCVS and a novel allocation model for the frame sampling rate. In addition, the authors define several indexes to assess the degree of flicker in decoded video. The experimental results demonstrate that the proposed algorithm reduces the degree of flicker in fast-motion sequences by 40-60% relative to the state-of-the-art architecture, while also outperforming other adaptive GOP selection strategies.

Vethamuthu Nesamony Manju;Alfred Lenin Fred; "AC coefficient and K-means cuckoo optimisation algorithm-based segmentation and compression of compound images," vol.12(2), pp.218-225, 2 2018. Compound images are containing palletise regions including text or graphics and continuous tone images. The compression of compound images is a challenging function and which is complicated to achieve it without degrading the quality of the images. This document is mainly used to improve the compression ratio and an efficient segmentation method is created to separate the background image, text and graphics from the compound images for to make the compression independently. The segmentation is performed through AC coefficient-based segmentation method resulting in smooth and non-smooth regions. The non-smooth region is again segmented by means of K-means cuckoo optimisation algorithm. In the second phase, the segmented background image, text and graphics were compressed by means of arithmetic coder, Huffman coder and JPEG coder, respectively. This proposed technique is implemented in the working platform of MATLAB and the results were analysed.

Liangtian He;Yilun Wang; "Image smoothing via truncated <inline-formula><tex-math notation="TeX">$ell _0$</tex-math>0<inline-graphic xlink:href="IET-IPR.2017.0533.IM1.gif" /></inline-formula> gradient regularisation," vol.12(2), pp.226-234, 2 2018. Edge-preserving image smoothing aims at maintaining the fundamental constituents, i.e. salient edges of a given image, while removing the noise and insignificant details in the meantime. It is often employed at the pre-processing step of many image processing tasks, including reconstruction, segmentation, recognition and three-dimensional content generation, to name just a few. Recently, a sparse gradient counting scheme in an optimisation framework has attracted much attention, and it confines the discrete number of intensity changes among neighbouring pixels. This ℓ0-regularised sparsity pursuit scheme performs favourably in a global optimisation manner. However, it often achieves unsatisfactory performance at diminishing trivial details and at smoothing discrete regions. In this study, a new image smoothing scheme with truncated0 regularisation is proposed, which is especially effective for sharpening critical edges. For objective evaluation of the smoothing performance, images are linearly quantised into several layers to generate the experimental images, then these quantised images are smoothed using several methods for reconstructing the smoothly changed shape and intensity of the original images. Compared with the ℓ0 smoothing scheme, extensive experimental results demonstrate that the proposed method performs much better at preserving main structures and removing trivial details.

Tao Ma; "Low-complexity and efficient image coder/decoder with quad-tree search model for embedded computing platforms," vol.12(2), pp.235-242, 2 2018. Among the existing image coding methods, set partition in hierarchy tree (SPIHT) becomes a favourable choice for the energy-constrained embedded computing system because of its simplicity and high coding efficiency. In this study, the authors presented a quad-tree search model which is able to provide the higher coding efficiency than the model used in SPIHT. By applying this model, the authors proposed a new image codec, which is able to surpass SPIHT by 0.2-0.5 dB in peak signal-to-noise ratio over various code rates, and its performance is even comparable to SPIHT with adaptive arithmetic code and JPEG2000. Also an experiment is conducted to show the complexity of the proposed codec is the same as SPIHT without the complicated arithmetic codes. This property is critically favoured in embedded processing communication systems where energy consumption and speed are priority concerns.

Randa Khemiri;Hassan Kibeya;Fatma Ezahra Sayadi;Nejmeddine Bahri;Mohamed Atri;Nouri Masmoudi; "Optimisation of HEVC motion estimation exploiting SAD and SSD GPU-based implementation," vol.12(2), pp.243-253, 2 2018. The new High-Efficiency Video Coding (HEVC) standard doubles the video compression ratio compared to the previous H.264/AVC at the same video quality and without any degradation. However, this important performance is achieved by increasing the encoder computational complexity. That's why HEVC complexity is a crucial subject. The most time consuming and the most intensive computing part of HEVC is the motion estimation based principally on the sum of absolute differences (SAD) or the sum of square differences (SSD) algorithms. For these reasons, the authors proposed an implementation of these algorithms on a low cost NVIDIA GPU (graphics processing unit) using the Fermi architecture developed with Compute Unified Device Architecture language. The proposed algorithm is based on the parallel-difference and the parallel-reduction process. The investigational results show a significant speed-up in terms of execution time for most 64 × 64 pixel blocks. In fact, the proposed parallel algorithm permits a significant reduction in the execution time that reaches up to 56.17 and 30.4%, compared to the CPU, for SAD and SSD algorithms, respectively. This improvement proves that parallelising the algorithm with the new proposed reduction process for the Fermi-GPU generation leads to better results. These findings are based on a static study that determines the PU percentage utilisation for each dimension in the HEVC. This study shows that the larger PUs are the most utilised in temporal levels 3 and 4, which attain 84.56% for class E. This improvement is accompanied by an average peak signal-to-noise ratio loss of 0.095 dB and a decrease of 0.64% in terms of BitRate.

Zhenghua Huang;Qian Li;Tianxu Zhang;Nong Sang;Hanyu Hong; "Iterative weighted sparse representation for X-ray cardiovascular angiogram image denoising over learned dictionary," vol.12(2), pp.254-261, 2 2018. Non-local self-similar patch-based denoising techniques have been viewed as the most popular denoising approaches in computer vision. This study has proposed a novel iterative weighted sparse representation (IWSR) scheme for X-ray cardiovascular angiogram image denoising. The main procedures of this scheme include four parts. First, a maximum a posterior (MAP) distribution by the Bayes' theory is adopted to simultaneously estimate the estimated image and sparse representation with different Gaussian distributions approximating to likelihood prior, non-local self-similar patch prior and sparse representation prior. Second, the MAP problem is converted to minimise an energy function using the logarithmic transformation. Third, the function is efficiently solved by the single and effective alternating directions method of multipliers algorithm along with singular value decomposition (SVD) algorithm. Finally, owing to learned dictionary by K-SVD algorithm, the qualitative and quantitative results of widely synthetic experiments demonstrate that the proposed IWSR denoising method performs effectively and can obtain competitive denoising performance and high-quality images compared with those advanced denoising methods. The results of extensive experiments on clinical X-ray angiogram images further illustrate that the IWSR method performs well on noise reduction and vascular structures including edges and capillaries preservation, integral cardiovascular trees of which are beneficial for clinicians to diagnose and analyse cardiovascular diseases.

Xu Guanlei;Wang Xiaotong;Zhou Lijia;Xu Xiaogang; "Image decomposition and texture analysis via combined bi-dimensional Bedrosian's principles," vol.12(2), pp.262-273, 2 2018. Image decomposition is an important issue in image processing. The existing approaches including the bi-dimensional empirical mode decomposition (BEMD) still fail to separate monocomponents in multicomponents in many cases. To solve this problem, this study proposes a new image decomposition method based on the new derived combined bi-dimensional Bedrosian's principle that has not been reported anywhere else for image processing. First, this study investigates a few bi-dimensional Bedrosian's principles according to the bi-dimensional Hilbert transforms. Second, based on the derived bi-dimensional Bedrosian's principles and the original multicomponents, the authors provide the combined bi-dimensional Bedrosian's principle and the assisted components obtained through projections via optimisation so that these monocomponents in multicomponents can be separated in the case that the existing methods fail. Third, an iterative image decomposition method is proposed via the above principles to decompose the multicomponent image into true monocomponents. The proposed method can solve the problems caused by the cross-angle and amplitude ratio and frequency ratio between these components that BEMD fails to solve. Also, the phase and amplitude are estimated for texture analysis after the decomposition is demonstrated. Experiments are shown to support the proposed methods.

Mubeen Ghafoor;Shahzaib Iqbal;Syed Ali Tariq;Imtiaz A. Taj;Noman M. Jafri; "Efficient fingerprint matching using GPU," vol.12(2), pp.274-284, 2 2018. Graphical processing unit (GPU) has proven a beneficial tool in handling computationally intensive algorithms, by introducing massive parallelism in the calculations. In this study, an effective and low-cost fingerprint identification (FI) solution is proposed that can exploit the parallel computational power of GPU proficiently. It is achieved by mapping a generalised minutia neighbour-based novel encoding and matching algorithm on low-cost GPU technology. The proposed solution achieves high accuracy in comparison with two open source matchers and it is shown to be scalable by comparing matching performance on different GPUs. The proposed GPU implementation employs multithreading and loop unrolling, which minimises the use of nested loops and avoids sequential matching of encoded minutia features. After a thorough and careful designing of data structures, memory transfers and computations, a GPU-based fingerprint matching system is developed. It achieves on average 50,196 fingerprint matches per second on a single GPU. As compared to the sequential central processing unit implementation, the proposed system achieves a speed up of around 92 times, while maintaining the accuracy. The proposed system with matcher integrated on GPU can be considered as a good, low-cost, robust and efficient solution for large-scale applications of automated FI systems.

Noor Badshah;Hassan Shah; "Model for smoothing and segmentation of texture images using <inline-formula><tex-math notation="TeX">$L_0$</tex-math>L0<inline-graphic xlink:href="IET-IPR.2017.0136.IM1.gif" /></inline-formula> norm," vol.12(2), pp.285-291, 2 2018. Segmentation of texture images is always a challenging problem in image processing. The authors propose a novel model for segmentation of texture images based on L0 gradient norm. The model will do smoothing of texture in image and segmentation jointly. It is well known that L0 gradient norm smooths the image and preserve the edges. Keeping this in view, the proposed model is using L0 gradient norm for smoothing of texture in image and Chan-Vese energy for segmentation. For fast and efficient solution of the model, the authors use alternating minimisation algorithm. Experimental results of their proposed model, which are compared with well-known (state of the art) existing models, validate better performance of the proposed model.

Guojia Hou;Zhenkuan Pan;Baoxiang Huang;Guodong Wang;Xin Luan; "Hue preserving-based approach for underwater colour image enhancement," vol.12(2), pp.292-298, 2 2018. In this study, a novel underwater colour image enhancement approach based on hue preserving is presented by combining hue-saturation-intensity (HSI) and HS-value (HSV) colour models. In this study, the proposed wavelet-domain filtering (WDF) and constrained histogram stretching (CHS) algorithms are operated on HSI and HSV colour models, respectively. The degraded image is first converted from red-green-blue colour model into the HSI colour model, wherein the hue component H is preserved and WDF algorithm is executed on the S and I components. Similarly, the image is further converted into the HSV colour model, wherein H component is kept invariant as well and CHS algorithm is applied on the S and V components. The authors' key contribution is that the H preserving method can improve image quality in terms of contrast, colour rendition, non-uniform illumination, and denoising. In addition, experimental results show that the proposed approach outperforms several other state-of-the-art algorithms.

IEEE Transactions on Signal Processing - new TOC (2018 February 15) [Website]

Panagiotis A. Traganitis;Georgios B. Giannakis; "Sketched Subspace Clustering," vol.66(7), pp.1663-1675, April1, 1 2018. The immense amount of daily generated and communicated data presents unique challenges in their processing. Clustering, the grouping of data without the presence of ground-truth labels, is an important tool for drawing inferences from data. Subspace clustering (SC) is a relatively recent method that is able to successfully classify nonlinearly separable data in a multitude of settings. In spite of their high clustering accuracy, SC methods incur prohibitively high computational complexity when processing large volumes of high-dimensional data. Inspired by random sketching approaches for dimensionality reduction, the present paper introduces a randomized scheme for SC, termed Sketch-SC, tailored for large volumes of high-dimensional data. Sketch-SC accelerates the computationally heavy parts of state-of-the-art SC approaches by compressing the data matrix across both dimensions using random projections, thus enabling fast and accurate large-scale SC. Performance analysis as well as extensive numerical tests on real data corroborate the potential of Sketch-SC and its competitive performance relative to state-of-the-art scalable SC approaches.

Raja Giryes;Yonina C. Eldar;Alex M. Bronstein;Guillermo Sapiro; "Tradeoffs Between Convergence Speed and Reconstruction Accuracy in Inverse Problems," vol.66(7), pp.1676-1690, April1, 1 2018. Solving inverse problems with iterative algorithms is popular, especially for large data. Due to time constraints, the number of possible iterations is usually limited, potentially affecting the achievable accuracy. Given an error one is willing to tolerate, an important question is whether it is possible to modify the original iterations to obtain faster convergence to a minimizer achieving the allowed error without increasing the computational cost of each iteration considerably. Relying on recent recovery techniques developed for settings in which the desired signal belongs to some low-dimensional set, we show that using a coarse estimate of this set may lead to faster convergence at the cost of an additional reconstruction error related to the accuracy of the set approximation. Our theory ties to recent advances in sparse recovery, compressed sensing, and deep learning. Particularly, it may provide a possible explanation to the successful approximation of the <inline-formula><tex-math notation="LaTeX">$ell _1$</tex-math> </inline-formula>-minimization solution by neural networks with layers representing iterations, as practiced in the learned iterative shrinkage-thresholding algorithm.

IEEE Signal Processing Letters - new TOC (2018 February 15) [Website]

Chongyi Li;Jichang Guo;Chunle Guo; "Emerging From Water: Underwater Image Color Correction Based on Weakly Supervised Color Transfer," vol.25(3), pp.323-327, March 2018. Underwater vision suffers from severe effects due to selective attenuation and scattering when light propagates through water. Such degradation not only affects the quality of underwater images, but limits the ability of vision tasks. Different from existing methods that either ignore the wavelength dependence on the attenuation or assume a specific spectral profile, we tackle color distortion problem of underwater images from a new view. In this letter, we propose a weakly supervised color transfer method to correct color distortion. The proposed method relaxes the need for paired underwater images for training and allows the underwater images being taken in unknown locations. Inspired by cycle-consistent adversarial networks, we design a multiterm loss function including adversarial loss, cycle consistency loss, and structural similarity index measure loss, which makes the content and structure of the outputs same as the inputs, meanwhile the color is similar to the images that were taken without the water. Experiments on underwater images captured under diverse scenes show that our method produces visually pleasing results, even outperforms the state-of-the-art methods. Besides, our method can improve the performance of vision tasks.

Xiao Fu;Kejun Huang;Nicholas D. Sidiropoulos; "On Identifiability of Nonnegative Matrix Factorization," vol.25(3), pp.328-332, March 2018. In this letter, we propose a new identification criterion that guarantees the recovery of the low-rank latent factors in the nonnegative matrix factorization (NMF) generative model, under mild conditions. Specifically, using the proposed criterion, it suffices to identify the latent factors if the rows of one factor are sufficiently scattered over the nonnegative orthant, while no structural assumption is imposed on the other factor except being full-rank. This is by far the mildest condition under which the latent factors are provably identifiable from the NMF model.

Jing Zhang;Xinhui Li;Peiguang Jing;Jing Liu;Yuting Su; "Low-Rank Regularized Heterogeneous Tensor Decomposition for Subspace Clustering," vol.25(3), pp.333-337, March 2018. This letter proposes a low-rank regularized heterogeneous tensor decomposition (LRRHTD) algorithm for subspace clustering, in which various constrains in different modes are incorporated to enhance the robustness of the proposed model. Specifically, due to the presence of noise and redundancy in the original tensor, LRRHTD seeks a set of orthogonal factor matrices for all but the last mode to map the high-dimensional tensor into a low-dimensional latent subspace. Furthermore, by imposing a low-rank constraint on the last mode, which is relaxed by using a nuclear norm, the lowest rank representation that reveals the global structure of samples is obtained for the purpose of clustering. We develop an effective algorithm based on the augmented Lagrange multiplier to optimize our model. Experiments on two public datasets demonstrate that our method reaches convergence within a small number of iterations and achieves promising results in comparison with the state of the arts.

Karan Nathwani;Emmanuel Vincent;Irina Illina; "DNN Uncertainty Propagation Using GMM-Derived Uncertainty Features for Noise Robust ASR," vol.25(3), pp.338-342, March 2018. The uncertainty decoding framework is known to improve the deep neural network (DNN)-based automatic speech recognition (ASR) performance in noisy environments. It operates by estimating the statistical uncertainty about the input features and propagating it to the output senone posteriors by sampling. Unfortunately, this approximate propagation scheme limits the performance improvement. In this letter, we exploit the fact that uncertainty propagation can be achieved in closed form for Gaussian mixture acoustic models (GMMs). We introduce new GMM-derived (GMMD) uncertainty features for the robust DNN-based acoustic model training and decoding. The GMMD features are computed as the difference between the GMM log-likelihoods obtained with versus without uncertainty. They are concatenated with conventional acoustic features and used as inputs to the DNN. We evaluate the resulting ASR performance on the CHiME-2 and CHiME-3 datasets. The proposed features are shown to improve the performance on both datasets, both for the conventional decoding and for the uncertainty decoding with different uncertainty estimation/propagation techniques.

Morteza Ashraphijuo;Vaneet Aggarwal;Xiaodong Wang; "On Deterministic Sampling Patterns for Robust Low-Rank Matrix Completion," vol.25(3), pp.343-347, March 2018. In this letter, we study the deterministic sampling patterns for the completion of low-rank matrix, when corrupted with a sparse noise, also known as robust matrix completion. We extend the recent results on the deterministic sampling patterns in the absence of noise based on the geometric analysis on the Grassmannian manifold. A special case where each column has a certain number of noisy entries is considered, where our probabilistic analysis performs very efficiently. Furthermore, assuming that the rank of the original matrix is not given, we provide an analysis to determine if the rank of a valid completion is indeed the actual rank of the data corrupted with sparse noise by verifying some conditions.

Subhadip Mukherjee;Chandra Sekhar Seelamantula; "Phase Retrieval From Binary Measurements," vol.25(3), pp.348-352, March 2018. We consider the problem of signal reconstruction from quadratic measurements that are encoded as <inline-formula> <tex-math notation="LaTeX">$+1$</tex-math></inline-formula> or <inline-formula><tex-math notation="LaTeX">$-1$ </tex-math></inline-formula> depending on whether they exceed a predetermined positive threshold or not. Binary measurements are fast to acquire and inexpensive in terms of hardware. We formulate the problem of signal reconstruction using a consistency criterion, wherein one seeks to find a signal that is in agreement with the acquired measurements. To enforce consistency, we construct a convex cost using a one-sided quadratic penalty and minimize it using an iterative accelerated projected gradient-descent technique. The projected gradient-descent (PGD) scheme reduces the cost function in each iteration, whereas incorporating momentum into PGD, notwithstanding the lack of such a descent property, exhibits faster convergence than PGD empirically. We refer to the resulting algorithm as binary phase retrieval (BPR). Considering additive white noise contamination prior to quantization, we also derive the Cramér–Rao Bound (CRB) for the binary encoding model. Experimental results demonstrate that the BPR algorithm yields a signal-to-reconstruction error ratio (SRER) of approximately 25 dB in the absence of noise. In the presence of noise prior to quantization, the SRER is within 2 to 3 dB of the CRB.

Mohammad Fayazur Rahaman;Mohammed Zafar Ali Khan; "Low-Complexity Optimal Hard Decision Fusion Under the Neyman–Pearson Criterion," vol.25(3), pp.353-357, March 2018. The design of the optimal nonrandomized hard decision fusion rule under the Neyman–Pearson criterion is known to be exponential in complexity. In this letter, we formulate a more generalized version of this problem called the “generalized decision fusion problem (GDFP)” and relate it to the classical <inline-formula> <tex-math notation="LaTeX">$text{0--1}$</tex-math></inline-formula> knapsack problem. Consequently, we show that the GDFP has a worst-case polynomial time solution. Numerical results are presented to verify the effectiveness of the proposed solution.

Ljubiša Stanković;Ervin Sejdić;Miloš Daković; "Vertex-Frequency Energy Distributions," vol.25(3), pp.358-362, March 2018. A vertex-varying spectral content on graphs challenges the assumption of vertex invariance and requires vertex-frequency representations for an adequate analysis. In this letter, we introduce a class of vertex-frequency energy distributions inspired by traditional time-frequency energy distributions. These newly introduced distributions do not use localization windows. Their efficiency in energy concentration is illustrated through examples.

Yuqi Li;Chong Wang;Jieyu Zhao; "Locally Linear Embedded Sparse Coding for Spectral Reconstruction From RGB Images," vol.25(3), pp.363-367, March 2018. Training-based spectral reconstruction is an efficient, inexpensive technique to recover spectral images from the RGB images captured by trichromatic cameras. Existing methods handle training samples individually without any consideration of local spatial and spectral correlations between samples, which results in high metamerism and inaccurate reconstruction. In this letter, we exploit for the first time the concept of spectral image reconstruction from RGB images with both chromatic and texture priors. We reduce redundancy of the sample set by applying a volume maximization based selection strategy. Taking advantage of the local linearity and sparsity of spectra in dictionary learning, we propose a locally linear embedded sparse reconstruction method taking into account both RGB values of pixels and the features of patch texture. Experimental results show that our method is significantly more accurate than the state-of-the-art methods.

Sahar Hashemgeloogerdi;Mark F. Bocko; "Inherently Stable Weighted Least-Squares Estimation of Common Acoustical Poles With the Application in Feedback Path Modeling Utilizing a Kautz Filter," vol.25(3), pp.368-372, March 2018. In adaptive feedback cancellation, the feedback path must be modeled precisely using as few adaptive parameters as possible to reduce computational complexity and enable rapid convergence. To reduce the number of adaptive parameters, the feedback path is usually modeled with a transfer function composed of an invariant component and a variable component using all-pole, all-zero, and pole-zero filters. These filters may be inefficient, requiring a large number of parameters for their specification, particularly in reverberant environments. In this letter, we present a weighted least-squares algorithm to precisely estimate the common poles of feedback paths, and we then model the invariant basis functions employing a Kautz filter. The Kautz filter is defined by a set of the fixed poles and a corresponding set of tap output weights. The fixed poles are associated with prominent peaks that are common to the measured feedback path frequency responses. The algorithm guarantees unconditionally the stability of the estimated poles of the inferred model. The experimental results using measured acoustic feedback paths from a two-microphone, behind-the-ear hearing aid show that the proposed method provides an accurate model of the feedback paths for a variety of acoustic environments that were not employed in estimating the common poles.

Vijayaditya Peddinti;Yiming Wang;Daniel Povey;Sanjeev Khudanpur; "Low Latency Acoustic Modeling Using Temporal Convolution and LSTMs," vol.25(3), pp.373-377, March 2018. Bidirectional long short-term memory (BLSTM) acoustic models provide a significant word error rate reduction compared to their unidirectional counterpart, as they model both the past and future temporal contexts. However, it is nontrivial to deploy bidirectional acoustic models for online speech recognition due to an increase in latency. In this letter, we propose the use of temporal convolution, in the form of time-delay neural network (TDNN) layers, along with unidirectional LSTM layers to limit the latency to 200 ms. This architecture has been shown to outperform the state-of-the-art low frame rate (LFR) BLSTM models. We further improve these LFR BLSTM acoustic models by operating them at higher frame rates at lower layers and show that the proposed model performs similar to these mixed frame rate BLSTMs. We present results on the Switchboard 300 h LVCSR task and the AMI LVCSR task, in the three microphone conditions.

Jian Jiang;Gang Wang;K. C. Ho; "Accurate Rigid Body Localization via Semidefinite Relaxation Using Range Measurements," vol.25(3), pp.378-382, March 2018. In this letter, we propose a method to improve the localization accuracy of a rigid body using the range measurements between the anchors and several sensors mounted on the body. Rigid body localization aims to estimate the rotation matrix and position vector of the sensor array on the body and it is a nonlinear and challenging problem to solve. Through the application of semidefinite relaxation to the original maximum likelihood formulation, we can obtain a coarse estimate of the rotation matrix and position vector that is robust to measurement noise. We next apply orthogonalization and refinement to the estimate and gain back the performance loss from relaxation and approximation. Simulation results show that the proposed method can reach the Cramér–Rao lower bound performance even at high noise levels where existing estimators cannot.

Erik Agrell;Balázs Csébfalvi; "Multidimensional Sampling of Isotropically Bandlimited Signals," vol.25(3), pp.383-387, March 2018. A new lower bound on the average reconstruction error variance of multidimensional sampling and reconstruction is presented. It applies to sampling on arbitrary lattices in arbitrary dimensions, assuming a stochastic process with constant, isotropically bandlimited spectrum and reconstruction by the best linear interpolator. The lower bound is exact for any lattice at sufficiently high and low sampling rates. The two threshold rates where the error variance deviates from the lower bound gives two optimality criteria for sampling lattices. It is proved that at low rates, near the first threshold, the optimal lattice is the dual of the best sphere-covering lattice, which for the first time establishes a rigorous relation between optimal sampling and optimal sphere covering. A previously known result is confirmed at high rates, near the second threshold, namely, that the optimal lattice is the dual of the best sphere-packing lattice. Numerical results quantify the performance of various lattices for sampling and support the theoretical optimality criteria.

Fuwei Yang;Wenming Yang;Riqiang Gao;Qingmin Liao; "Discriminative Multidimensional Scaling for Low-Resolution Face Recognition," vol.25(3), pp.388-392, March 2018. Face images captured by surveillance videos usually have limited resolution. Due to resolution mismatch, it is hard to match high-resolution (HR) faces with low-resolution (LR) faces directly. Recently, multidimensional scaling (MDS) has been employed to solve the problem. In this letter, we proposed a more discriminative MDS method to learn a mapping matrix, which projects the HR images and LR images to a common subspace. Our method is discriminative since both interclass distances and intraclass distances are taken into consideration. We add an interclass constraint to enlarge the distances of different subjects in the subspace to ensure discriminability. Besides, we consider not only the relationship of HR–LR images, but also the relationship of HR–HR images and LR–LR images in order to preserve local consistency. Experimental results on FERET, Multi-PIE, and SCface databases demonstrate the effectiveness of our proposed approach.

Hyowon Ha;Tae-Hyun Oh;In So Kweon; "A Closed-Form Solution to Rotation Estimation for Structure from Small Motion," vol.25(3), pp.393-397, March 2018. The introduction of small motion techniques such as small angle rotation approximation has enabled the three-dimensional reconstruction from a small motion of a camera, so-called structure from small motion (SfSM). In this letter, we propose a closed-form solution dedicated to the rotation estimation problem in SfSM. We show that our method works with a minimal set of two points, and has mild conditions to produce a unique optimal solution in practice. Also, we introduce a three-step SfSM pipeline with better convergence and faster speed compared to the state-of-the-art SfSM approaches. The key to this improvement is the separated estimation of the rotation with the proposed two-point method in order to handle the bas-relief ambiguity that affects the convergence of the bundle adjustment. We demonstrate the effectiveness of our two-point minimal solution and the three-step SfSM approach in synthetic and real-world experiments under the small motion regime.

Xiangming Meng;Sheng Wu;Jiang Zhu; "A Unified Bayesian Inference Framework for Generalized Linear Models," vol.25(3), pp.398-402, March 2018. In this letter, we present a unified Bayesian inference framework for generalized linear models (GLM), which iteratively reduces the GLM problem to a sequence of standard linear model (SLM) problems. This framework provides new perspectives on some established GLM algorithms derived from SLM ones and also suggests novel extensions for some other SLM algorithms. Specific instances elucidated under such framework are the GLM versions of approximate message passing (AMP), vector AMP, and sparse Bayesian learning. It is proved that the resultant GLM version of AMP is equivalent to the well-known generalized approximate message passing. Numerical results for one-bit quantized compressed sensing demonstrate the effectiveness of this unified framework.

Yuheng Jia;Sam Kwong;Junhui Hou; "Semi-Supervised Spectral Clustering With Structured Sparsity Regularization," vol.25(3), pp.403-407, March 2018. Spectral clustering (SC) is one of the most widely used clustering methods. In this letter, we extend the traditional SC with a semi-supervised manner. Specifically, with the guidance of small amount of supervisory information, we build a matrix with anti-block-diagonal appearance, which is further utilized to regularize the product of the low-dimensional embedding and its transpose. Technically, we formulate the proposed model as a constrained optimization problem. Then, we relax it as a convex problem, which can be efficiently solved with the global convergence guaranteed via the inexact augmented Lagrangian multiplier method. Experimental results over four real-world datasets demonstrate that higher accuracy and normalized mutual information are achieved when compared with state-of-the-art methods.

Filip Tronarp;Ángel F. García-Fernández;Simo Särkkä; "Iterative Filtering and Smoothing in Nonlinear and Non-Gaussian Systems Using Conditional Moments," vol.25(3), pp.408-412, March 2018. This letter presents the development of novel iterated filters and smoothers that only require specification of the conditional moments of the dynamic and measurement models. This leads to generalizations of the iterated extended Kalman filter, the iterated extended Kalman smoother, the iterated posterior linearization filter, and the iterated posterior linearization smoother. The connections to the previous algorithms are clarified and a convergence analysis is provided. Furthermore, the merits of the proposed algorithms are demonstrated in simulations of the stochastic Ricker map where they are shown to have a similar or superior performance to competing algorithms.

Devendra Kumar Yadav;Gajraj Kuldeep;S. D. Joshi; "Ramanujan Sums as Derivatives and Applications," vol.25(3), pp.413-416, March 2018. In 1918, S. Ramanujan defined a family of trigonometric sums now known as Ramanujan sums. In this letter, we define a class of operators based on the Ramanujan sums termed here as Ramanujan class of operators. We then prove that these operators possess properties of first derivative and with a particular shift, of second derivative also. Applications of Ramanujan class of operators for edge detection and noise level estimation are also demonstrated.

Ziqi Zheng;Junyan Huo;Bingbing Li;Hui Yuan; "Fine Virtual View Distortion Estimation Method for Depth Map Coding," vol.25(3), pp.417-421, March 2018. In three-dimensional (3-D) video coding systems, depth maps represent the geometric information of a 3-D scene. Since depth maps are not displayed to viewers but to generate virtual views, the quality of depth maps needs to be measured by its effect on the virtual view quality, which is indicated by the virtual view distortion (VVD) in depth map coding. In this letter, a fine VVD estimation method is proposed based on the analysis of the VVD. Specifically, the depth distortion of the current pixel, the texture gradient of the colocated color video and the depth distortions of adjacent pixels are all taken into consideration to estimate the VVD accurately. Experimental results demonstrate that the proposed method can improve 13.1% bitrate saving compared with the sum of squared differences based depth distortion calculation method and can improve 1.2% bitrate saving compared with the VVD estimation method in three dimensional high efficiency video coding (3D-HEVC) reference software.

Sadjad Imani;Mohammad Mahdi Nayebi;Seyed Ali Ghorashi; "Colocated MIMO Radar SINR Maximization Under ISL and PSL Constraints," vol.25(3), pp.422-426, March 2018. This letter considers joint design of transmit waveform and receive filter in colocated multiple-input multiple-output (MIMO) radars in order to enhance target detection performance in the presence of signal-dependent interference. Here, signal-to-interference-plus-noise ratio (SINR) is maximized under the practical constraints of integrated sidelobe level (ISL) and peak sidelobe level (PSL) at the pulse compression output. We have also shown that the joint transmit signal and receive filter design can be formulated as a convex optimization problem, which can be efficiently solved using optimization toolbox (CVX). Simulation results show that our proposed algorithm is able to suppress interference efficiently under ISL and PSL constraints and get close to the SINR efficiency of unconstrained methods.

Yun Zhang;Bin Luo;Liangpei Zhang; "Permutation Preference Based Alternate Sampling and Clustering for Motion Segmentation," vol.25(3), pp.432-436, March 2018. In this letter, permutation preference is used to represent the data points for the linkage clustering to segment the tracking points belonging to different motions. In order to exclude the impact of outliers, an alternate sampling and clustering strategy is performed, that iteratively alternates between sampling the hypotheses within the clusters and clustering the points with the permutation preference. As a result, points with similar permutation preferences are sampled as the hypotheses, and outliers are effectively excluded, thus, making the preferences more distinguishable and improving the clustering. The iterative interaction between sampling and clustering results in a good convergent result. The proposed method obtains robust segmentation results with the Hopkins 155 dataset, which are better than the results obtained by the state-of-the-art methods.

Saehoon Kim;Seungjin Choi; "Sparse Circulant Binary Embedding: An Asymptotic Analysis," vol.25(3), pp.432-436, March 2018. Binary embedding refers to methods for embedding points in <inline-formula><tex-math notation="LaTeX">$mathbb {R}^{d}$</tex-math></inline-formula> into vertices in the Hamming cube of dimension <inline-formula> <tex-math notation="LaTeX">${mathcal{O}}(d)$</tex-math></inline-formula>, such that the normalized Hamming distance between the codes preserves a prespecified distance between vectors in the original space. A common approach to binary embedding is to use random projection, followed by one-bit quantization to produce binary codes. Of particular interest, in this letter, is sparse circulant binary embedding (SCBE), where a sparse random circulant matrix with random sampling at a rate of <inline-formula><tex-math notation="LaTeX"> $1-s$</tex-math></inline-formula> for <inline-formula><tex-math notation="LaTeX">$sin (0, 1)$</tex-math> </inline-formula> is used for random projection. The SCBE has the space complexity <inline-formula> <tex-math notation="LaTeX">${mathcal{O}}((1-s)d)$</tex-math></inline-formula>, while unstructured random projection has the space complexity <inline-formula><tex-math notation="LaTeX">${mathcal{O}}(d^2)$</tex-math></inline-formula>. We present an asymptotic analysis of SCBE, when <inline-formula><tex-math notation="LaTeX">$d$</tex-math> </inline-formula> approaches <inline-formula><tex-math notation="LaTeX">$infty$</tex-math></inline-formula>, showing that the performance of SCBE is comparable to that of binary embedding with unstructured random projection whilethe former has the space complexity <inline-formula><tex-math notation="LaTeX">${mathcal{O}}((1-s)d)$</tex-math> </inline-formula> and the time complexity <inline-formula><tex-math notation="LaTeX">${mathcal{O}}(dlog d)$</tex-math> </inline-formula> but the latter has both space and time complexities <inline-formula><tex-math notation="LaTeX"> ${mathcal{O}}(d^2)$</tex-math></inline-formula>.

Ihor Trots;Andrzej Nowicki;Michiel Postema; "Ultrasound Image Improvement by Code Bit Elongation," vol.25(3), pp.437-441, March 2018. This letter analyses the influence of the transducer bandwidth on the compression and the axial resolution of an ultrasound image. The distortion of an electrical signal visible in the final image is a major problem in ultrasonography. To solve this problem, the bit length in Golay-complementary sequences was elongated, narrowing the fractional bandwidth of the coded sequences. Therefore, more energy of the burst signal could be transferred through the ultrasound transducer. The experimental results obtained for transmission of the complementary Golay-coded sequences with two different bit lengths—one-cycle and two-cycles—have been compared, and the efficiency of the pulse compression and its influence on the axial resolution for two fractional bandwidths have been discussed. The results are presented for two transducers having a fractional bandwidth of 25% and 80% and operating at a 6-MHz frequency. The results obtained show that the elongation of the Golay single bit length (doubled in our case) compensate for the limited transducer bandwidth. Two-dimensional ultrasound images of a tissue-mimicking phantom are presented and demonstrate the benefits of the use of two-cycle bit length.

Shiliang Huang;Chensi Mao;Jinxu Tao;Zhongfu Ye; "A Novel Chinese Sign Language Recognition Method Based on Keyframe-Centered Clips," vol.25(3), pp.442-446, March 2018. Isolated sign language recognition (SLR) is a long-standing research problem. The existing methods consider inclusively ambiguous data to represent a sign and ignore the fact that only scarce key information can represent the sign efficiently since most information are redundant. Furthermore, inclusion of redundant information may result in inefficiency and difficulty in modeling the long-term dependency for SLR. This letter delivers a novel sequence-to-sequence learning method based on keyframe centered clips (KCCs) for Chinese SLR. Different from conventional methods, only key information is considered to represent a sign significantly. The frames-to-word task is transformed into a KCCs-to-subwords task successfully, to allow for different attention in the input data. The empirical results of the proposed method outperform significantly the state-of-the-art SLR systems on our dataset containing 310 Chinese sign language words.

Igor Djurović; "QML-RANSAC Instantaneous Frequency Estimator for Overlapping Multicomponent Signals in the Time-Frequency Plane," vol.25(3), pp.447-451, March 2018. Effective quasi maximum likelihood – random samples consensus algorithm is proposed for the instantaneous frequency (IF) estimation of overlapping signals in the time-frequency (TF) plane. The proposed algorithm is tested on signal with five overlapping components. The IF estimation accuracy is excellent with signal-to-noise ratio threshold below 0 dB.

Minqiu Chen;Xingpeng Mao;Xiaozhuan Long;Liang Xin; "Underdetermined Direct Localization of Emitters Based on Spatio-Temporal Processing," vol.25(3), pp.452-456, March 2018. Without maintaining the internal constraint of the received data, conventional two-step passive localization methods are considered to be suboptimal. Following the thought of direct processing, a novel localization method based on the information of time-difference-of-arrival and angle-of-arrival (AOA) is proposed in this letter. Similar to other direct localization algorithms, the proposed method does not require the procedure of data association. By taking advantages of the spatio-temporal processing, the proposed method can handle the underdetermined scenario (i.e., the number of emitters exceeds the number of sensors from all the stations) without the prior knowledge about the number of emitters. Compared with the localization algorithms based on AOA only, the proposed method has superior performance on the condition of low signal-to-noise ratio and large bandwidth.

IEEE Journal of Selected Topics in Signal Processing - new TOC (2018 February 15) [Website]

* "Frontcover," vol.11(8), pp.C1-C1, Dec. 2017.* Presents the front cover for this issue of the publication.

* "IEEE Journal of Selected Topics in Signal Processing publication information," vol.11(8), pp.C2-C2, Dec. 2017.* Presents a listing of the editorial board, board of governors, current staff, committee members, and/or society editors for this issue of the publication.

* "Table of Contents," vol.11(8), pp.1235-1235, Dec. 2017.* Presents the table of contents for this issue of the publication.

* "Blank Page," vol.11(8), pp.B1233-B1233, Dec. 2017.* This page or pages intentionally left blank.

* "Blank Page," vol.11(8), pp.B1234-B1234, Dec. 2017.* This page or pages intentionally left blank.

Bhuvana Ramabhadran;Nancy F. Chen;Mary P. Harper;Brian Kingsbury;Kate Knill; "Introduction to the Special Issue on End-to-End Speech and Language Processing," vol.11(8), pp.1237-1239, Dec. 2017. The eleven papers in this special section focus on end-to-end speech and language processing (SLP) which is a series of sequence-to-sequence learning problems. Conventional SLP systems map input to output sequences through module-based architectures where each module is independently trained. These have a number of limitations including local optima, assumptions about intermediate models and features, and complex expert knowledge driven steps. It can be difficult for non-experts to use and develop new applications. Integrated End-to-End (E2E) systems aim to simplify the solution to these problems through a single network architecture to map an input sequence directly to the desired output sequence without the need for intermediate module representations. E2E models rely on flexible and powerful machine learning models such as recurrent neural networks. The emergence of models for end-to-end speech processing has lowered the barriers to entry into serious speech research. This special issue showcases the power of novel machine learning methods in end-to-end speech and language processing.

Shinji Watanabe;Takaaki Hori;Suyoun Kim;John R. Hershey;Tomoki Hayashi; "Hybrid CTC/Attention Architecture for End-to-End Speech Recognition," vol.11(8), pp.1240-1253, Dec. 2017. Conventional automatic speech recognition (ASR) based on a hidden Markov model (HMM)/deep neural network (DNN) is a very complicated system consisting of various modules such as acoustic, lexicon, and language models. It also requires linguistic resources, such as a pronunciation dictionary, tokenization, and phonetic context-dependency trees. On the other hand, end-to-end ASR has become a popular alternative to greatly simplify the model-building process of conventional ASR systems by representing complicated modules with a single deep network architecture, and by replacing the use of linguistic resources with a data-driven learning method. There are two major types of end-to-end architectures for ASR; attention-based methods use an attention mechanism to perform alignment between acoustic frames and recognized symbols, and connectionist temporal classification (CTC) uses Markov assumptions to efficiently solve sequential problems by dynamic programming. This paper proposes hybrid CTC/attention end-to-end ASR, which effectively utilizes the advantages of both architectures in training and decoding. During training, we employ the multiobjective learning framework to improve robustness and achieve fast convergence. During decoding, we perform joint decoding by combining both attention-based and CTC scores in a one-pass beam search algorithm to further eliminate irregular alignments. Experiments with English (WSJ and CHiME-4) tasks demonstrate the effectiveness of the proposed multiobjective learning over both the CTC and attention-based encoder-decoder baselines. Moreover, the proposed method is applied to two large-scale ASR benchmarks (spontaneous Japanese and Mandarin Chinese), and exhibits performance that is comparable to conventional DNN/HMM ASR systems based on the advantages of both multiobjective learning and joint decoding without linguistic resources.

Hao Tang;Liang Lu;Lingpeng Kong;Kevin Gimpel;Karen Livescu;Chris Dyer;Noah A. Smith;Steve Renals; "End-to-End Neural Segmental Models for Speech Recognition," vol.11(8), pp.1254-1264, Dec. 2017. Segmental models are an alternative to frame-based models for sequence prediction, where hypothesized path weights are based on entire segment scores rather than a single frame at a time. Neural segmental models are segmental models that use neural network-based weight functions. Neural segmental models have achieved competitive results for speech recognition, and their end-to-end training has been explored in several studies. In this work, we review neural segmental models, which can be viewed as consisting of a neural network-based acoustic encoder and a finite-state transducer decoder. We study end-to-end segmental models with different weight functions, including ones based on frame-level neural classifiers and on segmental recurrent neural networks. We study how reducing the search space size impacts performance under different weight functions. We also compare several loss functions for end-to-end training. Finally, we explore training approaches, including multistage versus end-to-end training and multitask training that combines segmental and frame-level losses.

Patrick Doetsch;Mirko Hannemann;Ralf Schlüter;Hermann Ney; "Inverted Alignments for End-to-End Automatic Speech Recognition," vol.11(8), pp.1265-1273, Dec. 2017. In this paper, we propose an inverted alignment approach for sequence classification systems like automatic speech recognition (ASR) that naturally incorporates discriminative, artificial-neural-network-based label distributions. Instead of aligning each input frame to a state label as in the standard hidden Markov model (HMM) derivation, we propose to inversely align each element of an HMM state label sequence to a segment-wise encoding of several consecutive input frames. This enables an integrated discriminative model that can be trained end-to-end from scratch or starting from an existing alignment path. The approach does not assume the usual decomposition into a separate (generative) acoustic model and a language model, and allows for a variety of model assumptions, including statistical variants of attention. Following our initial paper with proof-of-concept experiments on handwriting recognition, the focus of this paper was the investigation of integrated training and an inverted decoding approach, whereas the acoustic modeling still remains largely similar to standard hybrid modeling. We provide experiments on the CHiME-4 noisy ASR task. Our results show that we can reach competitive results with inverted alignment and decoding strategies.

Tsubasa Ochiai;Shinji Watanabe;Takaaki Hori;John R. Hershey;Xiong Xiao; "Unified Architecture for Multichannel End-to-End Speech Recognition With Neural Beamforming," vol.11(8), pp.1274-1288, Dec. 2017. This paper proposes a unified architecture for end-to-end automatic speech recognition (ASR) to encompass microphone-array signal processing such as a state-of-the-art neural beamformer within the end-to-end framework. Recently, the end-to-end ASR paradigm has attracted great research interest as an alternative to conventional hybrid paradigms with deep neural networks and hidden Markov models. Using this novel paradigm, we simplify ASR architecture by integrating such ASR components as acoustic, phonetic, and language models with a single neural network and optimize the overall components for the end-to-end ASR objective: generating a correct label sequence. Although most existing end-to-end frameworks have mainly focused on ASR in clean environments, our aim is to build more realistic end-to-end systems in noisy environments. To handle such challenging noisy ASR tasks, we study multichannel end-to-end ASR architecture, which directly converts multichannel speech signal to text through speech enhancement. This architecture allows speech enhancement and ASR components to be jointly optimized to improve the end-to-end ASR objective and leads to an end-to-end framework that works well in the presence of strong background noise. We elaborate the effectiveness of our proposed method on the multichannel ASR benchmarks in noisy environments (CHiME-4 and AMI). The experimental results show that our proposed multichannel end-to-end system obtained performance gains over the conventional end-to-end baseline with enhanced inputs from a delay-and-sum beamformer (i.e., BeamformIT) in terms of character error rate. In addition, further analysis shows that our neural beamformer, which is optimized only with the end-to-end ASR objective, successfully learned a noise suppression function.

Bo Wu;Kehuang Li;Fengpei Ge;Zhen Huang;Minglei Yang;Sabato Marco Siniscalchi;Chin-Hui Lee; "An End-to-End Deep Learning Approach to Simultaneous Speech Dereverberation and Acoustic Modeling for Robust Speech Recognition," vol.11(8), pp.1289-1300, Dec. 2017. We propose an integrated end-to-end automatic speech recognition (ASR) paradigm by joint learning of the front-end speech signal processing and back-end acoustic modeling. We believe that “only good signal processing can lead to top ASR performance” in challenging acoustic environments. This notion leads to a unified deep neural network (DNN) framework for distant speech processing that can achieve both high-quality enhanced speech and high-accuracy ASR simultaneously. Our goal is accomplished by two techniques, namely: (i) a reverberation-time-aware DNN based speech dereverberation architecture that can handle a wide range of reverberation times to enhance speech quality of reverberant and noisy speech, followed by (ii) DNN-based multicondition training that takes both clean-condition and multicondition speech into consideration, leveraging upon an exploitation of the data acquired and processed with multichannel microphone arrays, to improve ASR performance. The final end-to-end system is established by a joint optimization of the speech enhancement and recognition DNNs. The recent REverberant Voice Enhancement and Recognition Benchmark (REVERB) Challenge task is used as a test bed for evaluating our proposed framework. We first report on superior objective measures in enhanced speech to those listed in the 2014 REVERB Challenge Workshop on the simulated data test set. Moreover, we obtain the best single-system word error rate (WER) of 13.28% on the 1-channel REVERB simulated data with the proposed DNN-based pre-processing algorithm and clean-condition training. Leveraging upon joint training with more discriminative ASR features and improved neural network based language models, a low single-system WER of 4.46% is attained. Next, a new multi-channel-condition joint learning and testing scheme delivers a state-of-the-art WER of 3.76% on the 8-channel simulated data with a single ASR system. Finally, we also report on a prelim- nary yet promising experimentation with the REVERB real test data.

Panagiotis Tzirakis;George Trigeorgis;Mihalis A. Nicolaou;Björn W. Schuller;Stefanos Zafeiriou; "End-to-End Multimodal Emotion Recognition Using Deep Neural Networks," vol.11(8), pp.1301-1309, Dec. 2017. Automatic affect recognition is a challenging task due to the various modalities emotions can be expressed with. Applications can be found in many domains including multimedia retrieval and human-computer interaction. In recent years, deep neural networks have been used with great success in determining emotional states. Inspired by this success, we propose an emotion recognition system using auditory and visual modalities. To capture the emotional content for various styles of speaking, robust features need to be extracted. To this purpose, we utilize a convolutional neural network (CNN) to extract features from the speech, while for the visual modality a deep residual network of 50 layers is used. In addition to the importance of feature extraction, a machine learning algorithm needs also to be insensitive to outliers while being able to model the context. To tackle this problem, long short-term memory networks are utilized. The system is then trained in an end-to-end fashion where-by also taking advantage of the correlations of each of the streams-we manage to significantly outperform, in terms of concordance correlation coefficient, traditional approaches based on auditory and visual handcrafted features for the prediction of spontaneous and natural emotions on the RECOLA database of the AVEC 2016 research challenge on emotion recognition.

Tzeviya Fuchs;Joseph Keshet; "Spoken Term Detection Automatically Adjusted for a Given Threshold," vol.11(8), pp.1310-1317, Dec. 2017. Spoken term detection (STD) is the task of determining whether and where a given word or phrase appears in a given segment of speech. Algorithms for STD are often aimed at maximizing the gap between the scores of positive and negative examples. As such they are focused on ensuring that utterances where the term appears are ranked higher than utterances where the term does not appear. However, they do not determine a detection threshold between the two. In this paper, we propose a new approach for setting an absolute detection threshold for all terms by introducing a new calibrated loss function. The advantage of minimizing this loss function during training is that it aims at maximizing not only the relative ranking scores, but also adjusts the system to use a fixed threshold and thus maximizes the detection accuracy rates. We use the new loss function in the structured prediction setting and extend the discriminative keyword spotting algorithm for learning the spoken term detector with a single threshold for all terms. We further demonstrate the effectiveness of the new loss function by training a deep neural Siamese network in a weakly supervised setting for template-based STD, again with a single fixed threshold. Experiments with the TIMIT, Wall Street Journal (WSJ), and Switchboard corpora showed that our approach not only improved the accuracy rates when a fixed threshold was used but also obtained higher area under curve (AUC).

Batuhan Gündoğdu;Bolaji Yusuf;Murat Saraçlar; "Joint Learning of Distance Metric and Query Model for Posteriorgram-Based Keyword Search," vol.11(8), pp.1318-1328, Dec. 2017. In this paper, we propose a novel approach to keyword search (KWS) in low-resource languages, which provides an alternative method for retrieving the terms of interest, especially for the out of vocabulary (OOV) ones. Our system incorporates the techniques of query-by-example retrieval tasks into KWS and conducts the search by means of the subsequence dynamic time warping (sDTW) algorithm. For this, text queries are modeled as sequences of feature vectors and used as templates in the search. A Siamese neural network-based model is trained to learn a frame-level distance metric to be used in sDTW and the proper query model frame representations for this learned distance. Experiments conducted on Intelligence Advanced Research Projects Activity Babel Program's Turkish, Pashto, and Zulu datasets demonstrate the effectiveness of our approach. In each of the languages, the proposed system outperforms the large vocabulary continuous speech recognition (LVCSR) based baseline for OOV terms. Furthermore, the fusion of the proposed system with the baseline system provides an average relative actual term weighted value (ATWV) improvement of 13.9% on all terms and, more significantly, the fusion yields an average relative ATWV improvement of 154.5% on OOV terms. We show that this new method can be used as an alternative to conventional LVCSR-based KWS systems, or in combination with them, to achieve the goal of closing the gap between OOV and in-vocabulary retrieval performances.

Hongjie Chen;Cheung-Chi Leung;Lei Xie;Bin Ma;Haizhou Li; "Multitask Feature Learning for Low-Resource Query-by-Example Spoken Term Detection," vol.11(8), pp.1329-1339, Dec. 2017. We propose a novel technique that learns a low-dimensional feature representation from unlabeled data of a target language, and labeled data from a nontarget language. The technique is studied as a solution to query-by-example spoken term detection (QbE-STD) for a low-resource language. We extract low-dimensional features from a bottle-neck layer of a multitask deep neural network, which is jointly trained with speech data from the low-resource target language and resource-rich nontarget language. The proposed feature learning technique aims to extract acoustic features that offer phonetic discriminability. It explores a new way of leveraging cross-lingual speech data to overcome the resource limitation in the target language. We conduct QbE-STD experiments using the dynamic time warping distance of the multitask bottle-neck features between the query and the search database. The QbE-STD process does not rely on an automatic speech recognition pipeline of the target language. We validate the effectiveness of multitask feature learning through a series of comparative experiments.

Cristina España-Bonet;Ádám Csaba Varga;Alberto Barrón-Cedeño;Josef van Genabith; "An Empirical Analysis of NMT-Derived Interlingual Embeddings and Their Use in Parallel Sentence Identification," vol.11(8), pp.1340-1350, Dec. 2017. End-to-end neural machine translation has overtaken statistical machine translation in terms of translation quality for some language pairs, specially those with large amounts of parallel data. Besides this palpable improvement, neural networks provide several new properties. A single system can be trained to translate between many languages at almost no additional cost other than training time. Furthermore, internal representations learned by the network serve as a new semantic representation of words-or sentences-which, unlike standard word embeddings, are learned in an essentially bilingual or even multilingual context. In view of these properties, the contribution of the present paper is twofold. First, we systematically study the neural machine translation (NMT) context vectors, i.e., output of the encoder, and their power as an interlingua representation of a sentence. We assess their quality and effectiveness by measuring similarities across translations, as well as semantically related and semantically unrelated sentence pairs. Second, as extrinsic evaluation of the first point, we identify parallel sentences in comparable corpora, obtaining an F1=98.2% on data from a shared task when using only NMT context vectors. Using context vectors jointly with similarity measures F1 reaches 98.9%.

Kartik Audhkhasi;Andrew Rosenberg;Abhinav Sethy;Bhuvana Ramabhadran;Brian Kingsbury; "End-to-End ASR-Free Keyword Search From Speech," vol.11(8), pp.1351-1359, Dec. 2017. Conventional keyword search (KWS) systems for speech databases match the input text query to the set of word hypotheses generated by an automatic speech recognition (ASR) system from utterances in the database. Hence, such KWS systems attempt to solve the complex problem of ASR as a precursor. Training an ASR system itself is a time-consuming process requiring transcribed speech data. Our prior work presented an ASR-free end-to-end system that needed minimal supervision and trained significantly faster than an ASR-based KWS system. The ASR-free KWS system consisted of three subsystems. The first subsystem was a recurrent neural network based acoustic encoder that extracted a finite-dimensional embedding of the speech utterance. The second subsystem was a query encoder that produced an embedding of the input text query. The acoustic and query embeddings were input to a feedforward neural network that predicted whether the query occurred in the acoustic utterance or not. This paper extends our prior work in several ways. First, we significantly improve upon our previous ASR-free KWS results by nearly 20% relative through improvements to the acoustic encoder. Next, we show that it is possible to train the acoustic encoder on languages other than the language of interest with only a small drop in KWS performance. Finally, we attempt to predict the location of the detected keywords by training a location-sensitive KWS network.

* "List of reviewers," vol.11(8), pp.1360-1364, Dec. 2017.* The publication offers a note of thanks and lists its reviewers.

* "IEEE Journal of Selected Topics in Signal Processing information for authors," vol.11(8), pp.1365-1366, Dec. 2017.* These instructions give guidelines for preparing papers for this publication. Presents information for authors publishing in this journal.

* "2017 Index IEEE Journal of Selected Topics in Signal Processing Vol. 11," vol.11(8), pp.1367-1382, Dec. 2017.* Presents the 2017 subject/author index for this publication.

* "IEEE Signal Processing Society Information," vol.11(8), pp.C3-C3, Dec. 2017.* Presents a listing of the editorial board, board of governors, current staff, committee members, and/or society editors for this issue of the publication.

* "Blank Page," vol.11(8), pp.C4-C4, Dec. 2017.* This page or pages intentionally left blank.

IEEE Signal Processing Magazine - new TOC (2018 February 15) [Website]

* "Front Cover," vol.35(1), pp.C1-C1, Jan. 2018.* Presents the front cover for this issue of the publication.

* "ICIP CFP," vol.35(1), pp.C2-C2, Jan. 2018.* Advertisement.

* "Table of Contents," vol.35(1), pp.1-2, Jan. 2018.* Presents the table of contents for this issue of the publication.

* "Staff Listing," vol.35(1), pp.2-2, Jan. 2018.* Presents a listing of the editorial board, board of governors, current staff, committee members, and/or society editors for this issue of the publication.

Robert W. Heath; "Taking the Next Step for IEEE Signal Processing Magazine [From the Editor[Name:_blank]]," vol.35(1), pp.4-171, Jan. 2018. Presents the introductory editorial for this issue of the publication.

Rabab Ward; "Collaboration Empowers Innovation [President's Message[Name:_blank]]," vol.35(1), pp.5-6, Jan. 2018. Presents the President’s message for this issue of the publication.

* "SPS Announces 2018 Class of DLs and Creates New Distinguished Industry Speaker Program [Society News[Name:_blank]]," vol.35(1), pp.7-12, Jan. 2018.* Presents information on the SPS society Distinguished Lecturer Series in 2018.

* "2017 Members-at-Large and Regional Directors-at-Large Election Results [Society News[Name:_blank]]," vol.35(1), pp.12-12, Jan. 2018.* Presents a listing of SPS soceity 2017 Members-at-Large and Regional Directors.

John Edwards; "Signal Processing Powers Next-Generation Prosthetics: Researchers Investigate Techniques That Enable Artificial Limbs to Behave More Like Their Natural Counterparts [Special Reports[Name:_blank]]," vol.35(1), pp.13-16, Jan. 2018. Prosthetic limbs have improved significantly over the past several years, and signal processing has played a key role in allowing these devices to operate more smoothly and precisely on command. Now, researchers are taking the next step forward by using signal processing approaches and methods to develop prosthetics that not only function reliably and efficiently but give wearers more natural control over artificial arms, hands, and legs. Researchers at London's Imperial College, for instance, have developed a prosthetic arm sensor technology that detects signals transmitted by nerves in the spinal cord. To control the prosthetic, the wearer simply has to think about controlling a phantom arm and imagine a simple maneuver, such as pinching two fingers together. The sensor technology then interprets electrical signals sent from the spine and uses them as commands. Existing robotic prosthetic arms are controlled by having the wearer twitch remaining muscles in his or her shoulder or arm. The new approach detects signals from spinal motor neurons in parts of the body that were left undamaged by the amputation.

* "Errata," vol.35(1), pp.16-16, Jan. 2018.* Presents corrections to the paper, “Perfecting protection for interactive multimedia: A survey of forward errror correction for low-delay interactive applications,” (Badr, A. et al), IEEE Signal Process. Mag., vol. 34, no. 2, pp. 95–113, Mar. 2017.

Fatih Porikli;Shiguang Shan;Cees Snoek;Rahul Sukthankar;Xiaogang Wang; "Deep Learning for Visual Understanding: Part 2 [From the Guest Editors[Name:_blank]]," vol.35(1), pp.17-19, Jan. 2018. Visual perception is one of our most essential and fundamental abilities that enables us to make sense of what our eyes see and interpret the world that surrounds us. It allows us to function and, thus, our civilization to survive. No sensory loss is more debilitating than blindness as we are, above all, visual beings. Close your eyes for a moment after reading this sentence and try grabbing something in front of you, navigating your way in your environment, or just walking straight, reading a book, playing a game, or perhaps learning something new. Of course, please do not attempt to drive a vehicle. As you would realize again and appreciate profoundly, we owe so much to this amazing facility. It is no coincidence that most of the electrical activity in the human brain and most of its cerebral cortex is associated with visual understanding. Computer vision is the field of study that develops solutions for visual perception. In other words, it aims to make computers understand the seen data in the same way that human vision does. It incorporates several scientific disciplines such as signal processing, machine learning, applied mathematics, sensing, geometry, optimization, statistics, and data sciences to name a few.

Alice Lucas;Michael Iliadis;Rafael Molina;Aggelos K. Katsaggelos; "Using Deep Neural Networks for Inverse Problems in Imaging: Beyond Analytical Methods," vol.35(1), pp.20-36, Jan. 2018. Traditionally, analytical methods have been used to solve imaging problems such as image restoration, inpainting, and superresolution (SR). In recent years, the fields of machine and deep learning have gained a lot of momentum in solving such imaging problems, often surpassing the performance provided by analytical approaches. Unlike analytical methods for which the problem is explicitly defined and domain-knowledge carefully engineered into the solution, deep neural networks (DNNs) do not benefit from such prior knowledge and instead make use of large data sets to learn the unknown solution to the inverse problem. In this article, we review deep-learning techniques for solving such inverse problems in imaging. More specifically, we review the popular neural network architectures used for imaging tasks, offering some insight as to how these deep-learning tools can solve the inverse problem. Furthermore, we address some fundamental questions, such as how deeplearning and analytical methods can be combined to provide better solutions to the inverse problem in addition to providing a discussion on the current limitations and future directions of the use of deep learning for solving inverse problem in imaging.

Anurag Arnab;Shuai Zheng;Sadeep Jayasumana;Bernardino Romera-Paredes;Måns Larsson;Alexander Kirillov;Bogdan Savchynskyy;Carsten Rother;Fredrik Kahl;Philip H.S. Torr; "Conditional Random Fields Meet Deep Neural Networks for Semantic Segmentation: Combining Probabilistic Graphical Models with Deep Learning for Structured Prediction," vol.35(1), pp.37-52, Jan. 2018. Semantic segmentation is the task of labeling every pixel in an image with a predefined object category. It has numerous applications in scenarios where the detailed understanding of an image is required, such as in autonomous vehicles and medical diagnosis. This problem has traditionally been solved with probabilistic models known as conditional random fields (CRFs) due to their ability to model the relationships between the pixels being predicted. However, deep neural networks (DNNs) recently have been shown to excel at a wide range of computer vision problems due to their ability to automatically learn rich feature representations from data, as opposed to traditional handcrafted features. The idea of combining CRFs and DNNs have achieved state-of-the-art results in a number of domains. We review the literature on combining the modeling power of CRFs with the representation-learning ability of DNNs, ranging from early work that combines these two techniques as independent stages of a common pipeline to recent approaches that embed inference of probabilistic models directly in the neural network itself. Finally, we summarize future research directions.

Antonia Creswell;Tom White;Vincent Dumoulin;Kai Arulkumaran;Biswa Sengupta;Anil A. Bharath; "Generative Adversarial Networks: An Overview," vol.35(1), pp.53-65, Jan. 2018. Generative adversarial networks (GANs) provide a way to learn deep representations without extensively annotated training data. They achieve this by deriving backpropagation signals through a competitive process involving a pair of networks. The representations that can be learned by GANs may be used in a variety of applications, including image synthesis, semantic image editing, style transfer, image superresolution, and classification. The aim of this review article is to provide an overview of GANs for the signal processing community, drawing on familiar analogies and concepts where possible. In addition to identifying different methods for training and constructing GANs, we also point to remaining challenges in their theory and application.

Rajeev Ranjan;Swami Sankaranarayanan;Ankan Bansal;Navaneeth Bodla;Jun-Cheng Chen;Vishal M. Patel;Carlos D. Castillo;Rama Chellappa; "Deep Learning for Understanding Faces: Machines May Be Just as Good, or Better, than Humans," vol.35(1), pp.66-83, Jan. 2018. Recent developments in deep convolutional neural networks (DCNNs) have shown impressive performance improvements on various object detection/recognition problems. This has been made possible due to the availability of large annotated data and a better understanding of the nonlinear mapping between images and class labels, as well as the affordability of powerful graphics processing units (GPUs). These developments in deep learning have also improved the capabilities of machines in understanding faces and automatically executing the tasks of face detection, pose estimation, landmark localization, and face recognition from unconstrained images and videos. In this article, we provide an overview of deep-learning methods used for face recognition. We discuss different modules involved in designing an automatic face recognition system and the role of deep learning for each of them. Some open issues regarding DCNNs for face recognition problems are then discussed. This article should prove valuable to scientists, engineers, and end users working in the fields of face recognition, security, visual surveillance, and biometrics.

Junwei Han;Dingwen Zhang;Gong Cheng;Nian Liu;Dong Xu; "Advanced Deep-Learning Techniques for Salient and Category-Specific Object Detection: A Survey," vol.35(1), pp.84-100, Jan. 2018. Object detection, including objectness detection (OD), salient object detection (SOD), and category-specific object detection (COD), is one of the most fundamental yet challenging problems in the computer vision community. Over the last several decades, great efforts have been made by researchers to tackle this problem, due to its broad range of applications for other computer vision tasks such as activity or event recognition, content-based image retrieval and scene understanding, etc. While numerous methods have been presented in recent years, a comprehensive review for the proposed high-quality object detection techniques, especially for those based on advanced deep-learning techniques, is still lacking. To this end, this article delves into the recent progress in this research field, including 1) definitions, motivations, and tasks of each subdirection; 2) modern techniques and essential research trends; 3) benchmark data sets and evaluation metrics; and 4) comparisons and analysis of the experimental results. More importantly, we will reveal the underlying relationship among OD, SOD, and COD and discuss in detail some open questions as well as point out several unsolved challenges and promising future works.

Siqi Nie;Meng Zheng;Qiang Ji; "The Deep Regression Bayesian Network and Its Applications: Probabilistic Deep Learning for Computer Vision," vol.35(1), pp.101-111, Jan. 2018. Deep directed generative models have attracted much attention recently due to their generative modeling nature and powerful data representation ability. In this article, we review different structures of deep directed generative models and the learning and inference algorithms associated with the structures. We focus on a specific structure that consists of layers of Bayesian networks (BNs) due to the property of capturing inherent and rich dependencies among latent variables. The major difficulty of learning and inference with deep directed models with many latent variables is the intractable inference due to the dependencies among the latent variables and the exponential number of latent variable configurations. Current solutions use variational methods, often through an auxiliary network, to approximate the posterior probability inference. In contrast, inference can also be performed directly without using any auxiliary network to maximally preserve the dependencies among the latent variables. Specifically, by exploiting the sparse representation with the latent space, max-max instead of maxsum operation can be used to overcome the exponential number of latent configurations. Furthermore, the max-max operation and augmented coordinate ascent (AugCA) are applied to both supervised and unsupervised learning as well as to various inference. Quantitative evaluations on benchmark data sets of different models are given for both data representation and feature-learning tasks.

Yanwei Fu;Tao Xiang;Yu-Gang Jiang;Xiangyang Xue;Leonid Sigal;Shaogang Gong; "Recent Advances in Zero-Shot Recognition: Toward Data-Efficient Understanding of Visual Content," vol.35(1), pp.112-125, Jan. 2018. With the recent renaissance of deep convolutional neural networks (CNNs), encouraging breakthroughs have been achieved on the supervised recognition tasks, where each class has sufficient and fully annotated training data. However, to scale the recognition to a large number of classes with few or no training samples for each class remains an unsolved problem. One approach is to develop models capable of recognizing unseen categories without any training instances, or zero-shot recognition/learning. This article provides a comprehensive review of existing zero-shot recognition techniques covering various aspects ranging from representations of models, data sets, and evaluation settings. We also overview related recognition tasks including one-shot and open-set recognition, which can be used as natural extensions of zero-shot recognition when a limited number of class samples become available or when zero-shot recognition is implemented in a real-world setting. We highlight the limitations of existing approaches and point out future research directions in this existing new research area.

Yu Cheng;Duo Wang;Pan Zhou;Tao Zhang; "Model Compression and Acceleration for Deep Neural Networks: The Principles, Progress, and Challenges," vol.35(1), pp.126-136, Jan. 2018. In recent years, deep neural networks (DNNs) have received increased attention, have been applied to different applications, and achieved dramatic accuracy improvements in many tasks. These works rely on deep networks with millions or even billions of parameters, and the availability of graphics processing units (GPUs) with very high computation capability plays a key role in their success. For example, Krizhevsky et al. achieved breakthrough results in the 2012 ImageNet Challenge using a network containing 60 million parameters with five convolutional layers and three fully connected layers. Usually, it takes two to three days to train the whole model on the ImagetNet data set with an NVIDIA K40 machine. In another example, the top face-verification results from the Labeled Faces in the Wild (LFW) data set were obtained with networks containing hundreds of millions of parameters, using a mix of convolutional, locally connected, and fully connected layers. It is also very time-consuming to train such a model to obtain a reasonable performance. In architectures that only rely on fully connected layers, the number of parameters can grow to billions.

Deborah Cohen;Shahar Tsiper;Yonina C. Eldar; "Analog-to-Digital Cognitive Radio: Sampling, Detection, and Hardware," vol.35(1), pp.137-166, Jan. 2018. The radio spectrum is the radio-frequency (RF) portion of the electromagnetic spectrum. These spectral resources are traditionally allocated to licensed or primary users (PUs) by governmental organizations. As discussed in "Radio-Frequency Spectral Resources," most of the frequency bands are already allocated to one or more PUs. Consequently, new users cannot easily find free frequency bands. Spurred by the everincreasing demand from new wireless communication applications, this issue has become critical over the past few years.

Osvaldo Simeone; "Introducing Information Measures via Inference [Lecture Notes[Name:_blank]]," vol.35(1), pp.167-171, Jan. 2018. Information measures, such as the entropy and the Kullback-Leibler (KL) divergence, are typically introduced using an abstract viewpoint based on a notion of "surprise." Accordingly, the entropy of a given random variable (rv) is larger if its realization, when revealed, is on average more "surprising". The goal of this lecture note is to describe a principled and intuitive introduction to information measures that builds on inference, i.e., estimation and hypothesis testing. Specifically, entropy and conditional entropy measures are defined using variational characterizations that can be interpreted in terms of the minimum Bayes risk in an estimation problem. Divergence metrics are similarly described using variational expressions derived via mismatched estimation or binary hypothesis testing principles. The classical Shannon entropy and the KL divergence are recovered as special cases of more general families of information measures.

* "Calendar [Dates Ahead[Name:_blank]]," vol.35(1), pp.172-172, Jan. 2018.* Presents the SPS society calendar of upcoming events.

* "Errata," vol.35(1), pp.178-178, Jan. 2018.* Presents corrections to the paper, “Deep learning for image-to-text generation" (He, X. and Deng, L.), IEEE Signal Process. Mag., vol. 34, no. 6, pp. 109–116, Nov. 2017.

Li Deng; "Artificial Intelligence in the Rising Wave of Deep Learning: The Historical Path and Future Outlook [Perspectives[Name:_blank]]," vol.35(1), pp.180-177, Jan. 2018. Artificial intelligence (AI) is a branch of computer science and a technology aimed at developing the theories, methods, algorithms, and applications for simulating and extending human intelligence. Modern AI enables going from an old world-where people give computers rules to solve problems-to a new world-where people give computers problems directly and the machines learn how to solve them on their own using a set of algorithms. An algorithm is a self-contained sequence of instructions and actions to be performed by a computational machine. Starting from an initial state and initial input, the instructions describe computational steps, which, when executed, proceed through a finite number of well-defined successive states, eventually producing an output and terminating at a final ending state. AI algorithms are a rich set of algorithms used to perform AI tasks, notably those pertaining to perception and cognition that involve learning from data and experiences simulating human intelligence.

* "IEEE GlobalSIP," vol.35(1), pp.C3-C3, Jan. 2018.* Advertisement.

IET Signal Processing - new TOC (2018 February 15) [Website]

Wenpeng Zhang;Yaowen Fu;Lei Nie;Guanhua Zhao;Wei Yang;Jin Yang; "Parameter estimation of micro-motion targets for high-range-resolution radar using high-order difference sequence," vol.12(1), pp.1-11, 2 2018. Micro-range (m-R) signatures which are induced by micro-motion dynamics can be observed from range profiles, providing that the range resolution of radar is high enough. For real scenarios, micro-motion is often mixed with macro-motion (translation). To extract the micro-motion signatures, it is required to remove the macro-motion component. The widely employed range alignment technique fails for rigid-body targets with micro-motion, since the relative distances between different scattering centres on a rigid-body target are varying and it is unable to obtain a stable reference range profile. Thus, the extracted m-R signatures will be accompanied with residual macro-motion, which may lead to the degradation. However, this issue is often ignored in the research of m-R signatures extraction. In this work, by modelling the motions of scattering centres as the superimposition of a polynomial signal (represents macro-motion) and a sinusoidal signal (represents micro-motion), a micro-motion period estimation method based on high-order difference sequence is proposed. The property that the difference operation can decrease the order of polynomial signals while preserve sinusoidal signals with the same frequency enables the proposed method to extract m-R signatures in the presence of macro-motion. The effectiveness of the proposed method is validated by synthetic and measured radar data.

V.P. Ananthi;P. Balasubramaniam;P. Raveendran; "Impulse noise detection technique based on fuzzy set," vol.12(1), pp.12-21, 2 2018. In this study, a new fuzzy-based technique is introduced for denoising images corrupted by impulse noise. The proposed method is based on the intuitionistic fuzzy set (IFS), in which the degree of hesitation plays an important role. The degree of hesitation of the pixels is obtained from the values of memberships of the object and the background of the image. After minimising the obtained hesitation function, the IFS is constructed and the noisy pixels are detected outside the neighbourhood of mean intensity of the object and the background of an image. Denoised images are relatively analysed with five other methods: modified decision-based unsymmetric trimmed median filter, noise adaptive fuzzy switched median filter, adaptive fuzzy switching weighted average filter, adaptive weighted mean filter, iterative alpha trimmed mean filter. Performances of the proposed method along with these five state-of the-art methods are evaluated using a peak signal-to-noise ratio and error rate along with the time for computation. Experimentally, derived denoising method showed an improved performance than five other existing techniques in filtering noise in images due to the reduction of uncertainty while choosing the noisy pixels.

Lingfeng Liu;Shidi Hao;Jun Lin;Ze Wang;Xinyi Hu;Suoxia Miao; "Image block encryption algorithm based on chaotic maps," vol.12(1), pp.22-30, 2 2018. Advanced in unpredictability, ergodicity and sensitivity to initial conditions and parameters, chaotic maps are widely used in modern image encryption algorithms. In this study, the authors propose a novel image block encryption algorithm based on several widely used chaotic maps. In their algorithm, the image blocking method is variable, and both shuffling and substitution algorithms are adopted based on different chaotic maps. Several simulations are provided to evaluate the performances of this encryption scheme. The results demonstrate that the proposed algorithm is with high security level and fast encryption speed, which can be competitive with some other recently proposed image encryption algorithms.

Guojun Jiang;Xingpeng Mao;Yongtan Liu; "Reducing errors for root-MUSIC-based methods in uniform circular arrays," vol.12(1), pp.31-36, 2 2018. Root-MUSIC can be applied to uniform circular array (UCA) to achieve computationally efficient direction of arrival (DOA) estimation via beamspace transformation (BT). When the number of sensors of a UCA is small, the residual errors introduced by the BT will have significant values, resulting in performance degradation for DOA estimation. To solve this problem, an algorithm that enables the modification of the beamspace sample covariance matrix (BSCM) by considering the residual components is proposed. First, the residual components in the BSCM are calculated based on the initial DOAs estimated by the UCA root-MUSIC-based methods. Then the BSCM is modified by removing the undesirable terms. Finally, better DOA estimation performance is obtained by using the revised BSCM. The computational complexity and estimation error are derived. The significant advantages of the proposed algorithm are demonstrated by the simulation results.

Yong Yang;Dongling Zhang;Hua Peng; "Single-channel blind source separation for paired carrier multiple access signals," vol.12(1), pp.37-41, 2 2018. Paired carrier multiple access (PCMA) is one of the most common single-channel mixtures. It is still a great challenge to recover the transmitted bits from non-cooperative received PCMA signals, due to the high complexity of existing single-channel blind source separation (SCBSS) algorithms. Hence, a double-direction delayed-decision-feedback sequence estimation (DD-DDFSE) algorithm for non-causal high-order channel is proposed to realise the separation of PCMA signals. The proposed algorithm employs the Viterbi algorithm (VA) with low-order channel twice instead of conventional VA with high-order channel once. The symbols estimated by the initial VA are utilised to equalise the causal and non-causal taps of the channel beyond the trellis state in the second VA. The relationship between the decision-feedback error from the initial VA and the performance of the second VA is also derived. Compared with state-of-the-art maximum likelihood sequence estimation algorithms, DD-DDFSE algorithm can not only decrease the computation complexity but also improve the separation performance.

Yi Yu;Haiquan Zhao;Badong Chen; "Set-membership improved normalised subband adaptive filter algorithms for acoustic echo cancellation," vol.12(1), pp.42-50, 2 2018. In order to improve the performances of recently presented improved normalised subband adaptive filter (INSAF) and proportionate INSAF algorithms for highly noisy system, this study proposes their set-membership versions by exploiting the theory of set-membership filtering. Apart from obtaining smaller steady-state error, the proposed algorithms significantly reduce the overall computational complexity. In addition, to further improve the steady-state performance for the algorithms, their smooth variants are developed by using the smoothed absolute subband output errors to update the step sizes. Simulation results in the context of acoustic echo cancellation have demonstrated the superiority of the proposed algorithms.

Vindheshwari P. Singh;Ajit K. Chaturvedi; "Statistically robust transceiver design algorithms for relay aided multiple-input multiple-output interference systems," vol.12(1), pp.51-63, 2 2018. In this study, the authors consider robust transceiver design for amplify-and-forward relay aided multiple-input multiple-output interference systems assuming direct transmitter-receiver links and imperfect channel state information (CSI). The imperfect CSI of each link consists of the estimated channel and the covariance matrix of the channel estimation error. The authors address two transceiver optimisation problems: one is minimising the sum of averaged mean squared error (MSE) at all the receivers and the other is minimising the maximum averaged MSE among all the receivers, both of which are subject to power constraints at the transmitter and relay nodes. The formulated sum-MSE minimisation and max-MSE minimisation based optimisation problems are non-convex with matrix variables and therefore a globally optimal solution is difficult to obtain. To solve these non-convex optimisation problems, they develop sub-optimal iterative algorithms based on alternating minimisation approach to jointly optimise the precoding matrices at the transmitter and relay nodes and the receiver filter matrices. Simulation results demonstrate the effectiveness of the proposed algorithms and their improved performance against CSI uncertainties at similar computational cost as the existing non-robust designs.

Xingwang Li;Jingjing Li;Lihua Li;Liutong Du;Jin Jin;Di Zhang; "Performance analysis of cooperative small cell systems under correlated Rician/Gamma fading channels," vol.12(1), pp.64-73, 2 2018. Small cell networks (SCNs) have emerged as promising technologies to meet the data traffic demands for the future wireless communications. However, the benefits of SCNs are limited to their hard handovers between base stations (BSs). In addition, the interference is another challenging issue. To solve this problem, this study employs a cooperative transmission mechanism focusing on correlated Rician/Gamma fading channels with zero-forcing receivers. The analytical expressions for the achievable sum rate, symbol error rate and outage probability are derived, which are applicable to arbitrary Rician factors, correlation coefficients, the number of antennas, and remain tight across entire signal-to-noise ratios (SNRs). Asymptotic analyses at the high and low SNR regimes are carried out in order to further reveal the insights of the model parameters on the system performance. Monte-Carlo simulation results validate the correctness of their derivations. Numerical results indicate that the theoretical expressions provide sufficiently accurate approximation to simulated results.

Jesus Selva; "Efficient type-4 and type-5 non-uniform FFT methods in the one-dimensional case," vol.12(1), pp.74-81, 2 2018. The so-called non-uniform fast Fourier transform (NFFT) is a family of algorithms for efficiently computing the Fourier transform of finite-length signals, whenever the time or frequency grid is non-uniformly spaced. Among the five usual NFFT types, types 4 and 5 involve an inversion problem, and this makes them the most intensive computationally. The usual efficient methods for these last types are either based on a fast multipole (FM) or on an iterative conjugate gradient (CG) method. The purpose of this study is to present efficient methods for these type-4 and type-5 NFFTs in the one-dimensional case that just require three NFFTs of types 1 or 2 plus some additional fast Fourier transforms (FFTs). Fundamentally, they are based on exploiting the Lagrange formula structure. The proposed methods roughly provide a factor-ten improvement on the FM and CG alternatives in computational burden. The study includes several numerical examples in double precision, in which the proposed and the Gaussian elimination, CG and FM methods are compared, both in terms of round-off error and computational burden.

Tatiana Chakravorti;Rajesh Kumar Patnaik;Pradipta Kishore Dash; "Detection and classification of islanding and power quality disturbances in microgrid using hybrid signal processing and data mining techniques," vol.12(1), pp.82-94, 2 2018. This study presents multi-scale morphological gradient filter (MSMGF) and short-time modified Hilbert transform (STMHT) techniques, respectively, to detect and classify multiclass power system disturbances in a distributed generation (DG)-based microgrid environment. The non-stationary power signal samples measured near the target DG's are processed through the proposed MSMGF and STMHT techniques, respectively, and some computations over them generates the target parameter sets. Depending on the complexity of the overlapping in the target attribute values for different disturbance patterns, fuzzy judgment tree structure is incorporated for multiclass event classification, which proves to be robust for most of the classes. In this regard, an extensive simulation on the proposed microgrid models, subjected to a number of multiclass disturbances has been performed in MATLAB/Simulink environment. The faster execution, lower computational burden, superior efficiency as well as better accuracy in multiclass power system disturbance classification by the proposed judgment tree-based MSMGF and STMHT techniques, respectively, as compared to some of the conventional techniques, is significantly illustrated in the performance evaluation section. Further, as illustrated in this section, the real-time capability of the proposed techniques has been verified in the hardware environment, where the results shown are satisfactory.

Rahul Kumar Vijay;Satyasai Jagannath Nanda; "Tetra-stage cluster identification model to analyse the seismic activities of Japan, Himalaya and Taiwan," vol.12(1), pp.95-103, 2 2018. From the decades, due to the independent and Poisson nature of background seismicity, they are extensively used for hazard analysis, modelling of prediction phenomenon and also used for earthquake simulations. In this study, a tetra-stage cluster identification model is proposed for accurate estimation of background seismicity and triggered seismicity. The proposed method considers a seismic event's occurrence time, location, magnitude and depth information available in the given catalogue to classify the event as a background or aftershock. The model has flexible threshold parameters which can be tuned to a proper value according to the specific seismic zone to be analysed. It exploits the current seismic activities of the region by taking care the past samples of the region over last 25 years. The analyses of Japan, Himalaya and Taiwan catalogues are carried out using the proposed model. Superior results with the proposed model are achieved, compared with benchmark models by Nanda et al., Gardner-Knopoff and Uhrhammer et al. in terms of percentage of background seismicity, lambda plot and cumulative plot. The ergodicity present in the original seismic catalogue and catalogue after de-clustering are compared using Thirumalai-Mountain metric to justify the stationary and linearity.

Tang Tang;Lijuan Jia;Jian Lou;Ran Tao;Yue Wang; "Adaptive EIV-FIR filtering against coloured output noise by using linear prediction technique," vol.12(1), pp.104-112, 2 2018. The problem of finite impulse response (FIR) filtering in errors-in-variables (EIV) system is studied. Due to the input noise, traditional recursive least-squares (RLS) algorithms are biased in EIV system. Most existing bias-compensated approaches are proposed in the case that both the input-output noises are white Gaussian random processes. However, taking account of the situation where the output is corrupted by coloured noise, there are rare existing algorithms work well. Two bias-compensated RLS algorithms with acceptable computational complexity are proposed, which can obtain unbiased real-time filtering in non-stationary system when the input noise is white while the output noise is coloured. Under the assumption that the input signal is a coloured process, linear prediction technique is used to estimate the sample of the input signal. Exploiting the statistical properties of the cross-correlation function between the least-squares error and the forward/backward prediction error, the input noise variance can be estimated and the bias can be compensated. Simulation results illustrate the good performance of the proposed algorithms.

Guimei Zheng;Dong Zhang; "BOMP-based angle estimation with polarimetric MIMO radar with spatially spread crossed-dipole," vol.12(1), pp.113-118, 2 2018. For polarimetric multi-input multi-output (MIMO) radar with spatially spread crossed-dipole, this article studies the problem of joint direction of arrival (DOA) and polarisation parameter estimation based on block-orthogonal matching pursuit (BOMP) algorithm. First, the signal model of polarimetric MIMO radar with spatially spread crossed-dipole is established, and then the covariance matrix of the received data is calculated. Using the relationship between polarisation parameter and DOA in the crossed-dipole, sparse dictionary matrix is constructed within only DOA parameter and it will be translated into a block sparse problem. Then, fast BOMP algorithm is used to estimate their support positions and their amplitudes. Last, DOA estimation is calculated by support positions and polarisation parameter is estimated by the amplitudes of the support positions. The proposed algorithm has three advantages. One is that overcomplete dictionary is constructed within only the DOA, which has a small computational complexity. Another one is that the problem of strong mutual coupling among collocated crossed-dipole is solved by using the spatially spread crossed-dipole. The last one is that the DOA and polarisation estimations can pair automatically without any additional processing. Computer simulation results demonstrate the effectiveness of the proposed algorithm.

Ting-An Chang;Wei-Chen Liao;Jar-Ferr Yang; "Robust depth enhancement based on texture and depth consistency," vol.12(1), pp.119-128, 2 2018. With advances in three-dimension television (3DTV) technology, accurate depth information for 3DTV broadcasting has gained much attention recently. The depth map, either retrieved by stereo matching or captured by the RGB-D camera, is mostly with lower resolution and often with noisy or missing values than the texture frame. How to effectively utilise high-resolution texture image to enhance the corresponding depth map becomes an important and inevitable approach. In this study, the authors propose texture similarity-based hole filling, texture similarity-based depth enhancement and rotating counsel depth refinement to enhance the depth map. Thus, the proposed depth enhancement system could suppress the noise, fill the holes and sharpen the object edges simultaneously. Experimental results demonstrate that the proposed system provides a superior performance, especially around the object boundary comparing to the state-of-the-art depth enhancement methods.

Basheera M. Mahmmod;Abd Rahman bin Ramli;Sadiq H. Abdulhussain;Syed Abdul Rahman Al-Haddad;Wissam A. Jassim; "Signal compression and enhancement using a new orthogonal-polynomial-based discrete transform," vol.12(1), pp.129-142, 2 2018. Discrete orthogonal functions are important tools in digital signal processing. These functions received considerable attention in the last few decades. This study proposes a new set of orthogonal functions called discrete Krawtchouk-Tchebichef transform (DKTT). Two traditional orthogonal polynomials, namely, Krawtchouk and Tchebichef, are combined to form DKTT. The theoretical and mathematical frameworks of the proposed transform are provided. DKTT was tested using speech and image signals from a well-known database under clean and noisy environments. DKTT was applied in a speech enhancement algorithm to evaluate the efficient removal of noise from speech signal. The performance of DKTT was compared with that of standard transforms. Different types of distance (similarity index) and objective measures in terms of image quality, speech quality, and speech intelligibility assessments were used for comparison. Experimental tests show that DKTT exhibited remarkable achievements and excellent results in signal compression and speech enhancement. Therefore, DKTT can be considered as a new set of orthogonal functions for futuristic applications of signal processing.

IEEE Transactions on Geoscience and Remote Sensing - new TOC (2018 February 15) [Website]

* "[Front cover[Name:_blank]]," vol.56(2), pp.C1-C1, Feb. 2018.* Presents the front cover for this issue of the publication.

* "IEEE Transactions on Geoscience and Remote Sensing publication information," vol.56(2), pp.C2-C2, Feb. 2018.* Provides a listing of current staff, committee members and society officers.

* "Table of contents," vol.56(2), pp.609-1216, Feb. 2018.* Presents the table of contents for this issue of the periodical.

Feng Wang;Feng Xu;Ya-Qiu Jin; "Three-Dimensional Reconstruction From a Multiview Sequence of Sparse ISAR Imaging of a Space Target," vol.56(2), pp.611-620, Feb. 2018. To monitor a space target, 3-D reconstruction from a multiview sequence of the inverse synthetic aperture radar (ISAR) imaging is developed. Scattering of a complex electric-large target, e.g., the ENVISAT satellite model, is numerically calculated, and multiview 2-D ISAR imaging can be simulated. Under the sparse sampling ISAR imaging via compressed sensing, the Kanade-Lucas–Tomasi feature tracker is applied to extraction of target feature points. Then, using the orthographic factorization method, 3-D reconstruction of those feature points is produced. A simple hexagonal frustum is first tested for the feasibility analysis. Two sequences of multiview ISAR imaging, one is the ENVISAT model and another real measurements of a space shuttle, are then presented for 3-D reconstruction. Furthermore, a complex multistructure model of the International Space Station is also studied from multiview ISAR imaging under different sparse sampling rates. All results demonstrate good feasibility of the 3-D reconstruction for those target components, e.g., solar panel and antenna.

Yuebin Wang;Liqiang Zhang;Xiaohua Tong;Feiping Nie;Haiyang Huang;Jie Mei; "LRAGE: Learning Latent Relationships With Adaptive Graph Embedding for Aerial Scene Classification," vol.56(2), pp.621-634, Feb. 2018. The performance of scene classification relies heavily on the spatial and structural features that are extracted from high spatial resolution remote-sensing images. Existing approaches, however, are limited in adequately exploiting latent relationships between scene images. Aiming to decrease the distances between intraclass images and increase the distances between interclass images, we propose a latent relationship learning framework that integrates an adaptive graph with the constraints of the feature space and label propagation for high-resolution aerial image classification. To describe the latent relationships among scene images in the framework, we construct an adaptive graph that is embedded into the constrained joint space for features and labels. To remove redundant information and improve the computational efficiency, subspace learning is introduced to assist in the latent relationship learning. To address out-of-sample data, linear regression is adopted to project the semisupervised classification results onto a linear classifier. Learning efficiency is improved by minimizing the objective function via the linearized alternating direction method with an adaptive penalty. We test our method on three widely used aerial scene image data sets. The experimental results demonstrate the superior performance of our method over the state-of-the-art algorithms in aerial scene image classification.

Pietro Guccione;Michele Scagliola;Davide Giudici; "Low-Frequency SAR Radiometric Calibration and Antenna Pattern Estimation by Using Stable Point Targets," vol.56(2), pp.635-646, Feb. 2018. In this paper, the synthetic aperture radar (SAR) calibration for low-frequency missions by means of stable point targets is presented. Calibration at low frequency involves the absolute radiometric calibration, the antenna pattern and pointing characterization and validation, and the distortion system parameters' estimation. The use of traditional instrumentation, such as a polarimetric active radar calibrator, a corner reflector, or an active transponder, may be costly and can reduce the time the instrument is used for operational acquisitions. The purpose of this paper is to evaluate the potentiality in calibration of point targets for which the radar cross section and the time stability have been characterized. Given a calibration site, once that a set of the stable point targets have been detected by the analysis of an interferometric stack of SAR acquisitions, they may be used as passive calibrators for the validation of radiometry, elevation antenna pattern, and pointing estimation. We show that, although less targets are expected to be found in P- or L-band than in C- or X-band, a sufficient amount (about 250 targets per acquisition) can provide an accuracy in antenna pattern estimation of about 0.04 dB, if the target accuracy is 0.1 dB at 1σ.

Sina Soltani;Mojtaba Kordestani;Paknoosh Karim Aghaee;Mehrdad Saif; "Improved Estimation for Well-Logging Problems Based on Fusion of Four Types of Kalman Filters," vol.56(2), pp.647-654, Feb. 2018. The concept of information fusion has gained a widespread interest in many fields due to its complementary properties. It makes systems more robust against uncertainty. This paper presents a new approach for the well-logging estimation problem by using a fusion methodology. The natural gamma-ray tool (NGT) is considered as an important instrument in the well logging. The NGT detects changes in natural radioactivity emerging from the variations in concentrations of micronutrients as uranium (U), thorium (Th), and potassium (K). The main goal of this paper is to have precise estimation of the concentrations of <inline-formula> <tex-math notation="LaTeX">$U$ </tex-math></inline-formula>, <inline-formula> <tex-math notation="LaTeX">$Th$ </tex-math></inline-formula>, and <inline-formula> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula>. Four types of Kalman filters are designed to estimate the elements using the NGT sensor. Then, a fusion of the Kalman filters is utilized into an integrated framework by an ordered weighted averaging (OWA) operator to enhance the quality of the estimations. A real covariance of the output error based on the innovation matrix is utilized to design weighting factors for the OWA operator. The simulation studies indicate not only a reliable performance of the proposed method compared with the individual Kalman filters but also a better response in contrast with previous fusion methodologies.

David M. Le Vine;Saji Abraham; "Faraday Rotation Correction for SMAP and Soil Moisture Retrieval," vol.56(2), pp.655-668, Feb. 2018. Faraday rotation can be significant at L-band and needs to be considered in remote sensing from space using the spectrum window at 1.413 GHz protected for passive observations. This is especially so for a conical scanner such as SMAP because the variation of the rotation angle with position around the scan is of the same order of magnitude as the change with geographic position as the sensor travels in its orbit around the globe. Furthermore, the angle retrieved in situ by the radiometer is particularly noisy over land raising additional issues for remote sensing of soil moisture. Research is reported here assessing the magnitude of the problem and suggesting an approach for treating Faraday rotation in the context of remote sensing of soil moisture with a conical scanner like SMAP.

Xiang Xu;Jun Li;Shutao Li; "Multiview Intensity-Based Active Learning for Hyperspectral Image Classification," vol.56(2), pp.669-680, Feb. 2018. In remote sensing image classification, active learning aims to learn a good classifier as best as possible by choosing the most valuable (informative and representative) training samples. Multiview is a concept that regards analyzing the same object from multiple different views. Generally, these views show diversity and complementarity of features. In this paper, we propose a new multiview active learning (MVAL) framework for hyperspectral image classification. First, we generate multiple views by extracting different attribute components from the same image data. Specifically, we adopt the multiple morphological component analysis to decompose the original image into multiple pairs of attribute components, including content, coarseness, contrast, and directionality, and the smooth component from each pair is chosen as one single view. Second, we construct two multiview intensity-based query strategies for active learning. On the one hand, we exploit the intensity differences of multiple views along with the samples’ uncertainty to choose the most informative candidates. On the other hand, we consider the clustering distribution of all unlabeled samples, and query the most representative candidates in addition to the highly informative ones. Our experiments are performed on four benchmark hyperspectral image data sets. The obtained results show that the proposed MVAL framework can lead to better classification performance than the traditional, single-view active learning schemes. In addition, compared with the conventional disagree-based MVAL scheme, the proposed query selection strategies show competitive classification accuracy.

C. Hakan Arslan;Kültegin Aydin;Julio V. Urbina;Lars Dyrud; "Satellite-Link Attenuation Measurement Technique for Estimating Rainfall Accumulation," vol.56(2), pp.681-693, Feb. 2018. A technique using satellite-link signal attenuation measurements for estimating rainfall accumulation along the link path is evaluated. Power law relationships between attenuation rate A and rainfall rate R are used to estimate R and rainfall accumulation with a satellite link operating at Ku-band (12.3 GHz). Polarimetric radar measurements obtained from a National Weather Service Weather Surveillance Radar-1988 Doppler system near State College, Pennsylvania, are utilized to provide a comparison of rainfall accumulation estimates. A tipping-bucket rain gauge, colocated with the satellite receiver, is also used for comparison. A method based on bit error ratio measurements for the satellite link is used to identify periods of rain during which the rainfall rate is estimated from signal attenuation measurements. The effective rain height used in converting the attenuation rate along the link path into the rainfall rate is estimated from polarimetric radar observations. The Ku-band link is not very sensitive to light rain below 1.5 mm/h. Rainfall accumulation estimates obtained for 11 different days using satellite link attenuation show good comparisons with radar (within 19%) for accumulations greater than 6 mm and not so good (within 43%) for accumulations below 3 mm. The results presented in this paper show that using satellite-link attenuation measurements to estimate rainfall accumulations is a promising technique that requires further testing and refinement.

Bin Yang;Bin Wang;Zongmin Wu; "Nonlinear Hyperspectral Unmixing Based on Geometric Characteristics of Bilinear Mixture Models," vol.56(2), pp.694-714, Feb. 2018. Recently, many nonlinear spectral unmixing algorithms that use various bilinear mixture models (BMMs) have been proposed. However, the high computational complexity and intrinsic collinearity between true endmembers and virtual endmembers considerably decrease these algorithms' unmixing performances. In this paper, we come up with a novel abundance estimation algorithm based on the BMMs. Motivated by BMMs' geometric characteristics that are related to collinearity, we conduct a unique nonlinear vertex p to replace all the virtual endmembers. Unlike the virtual endmembers, this vertex p actually works as an additional true endmember that gives affine representations of pixels with other true endmembers. When the pixels' normalized barycentric coordinates with respect to true endmembers are obtained, they will be directly projected to be their approximate linear mixture components, which removes the collinearity effectively and enables further linear spectral unmixing. After that, based on the analysis of projection bias, two strategies using the projected gradient algorithm and a traditional linear spectral unmixing algorithm, respectively, are provided to correct the bias and estimate more accurate abundances. The experimental results on simulated and real hyperspectral data show that the proposed algorithm performs better compared with both traditional and state-of-the-art spectral unmixing algorithms. Both the unmixing accuracy and speed have been improved.

Huajian Xu;Zhiwei Yang;Min Tian;Yongyan Sun;Guisheng Liao; "An Extended Moving Target Detection Approach for High-Resolution Multichannel SAR-GMTI Systems Based on Enhanced Shadow-Aided Decision," vol.56(2), pp.715-729, Feb. 2018. This paper develops a framework based on enhanced shadow-aided decision for multichannel synthetic aperture radar-based ground moving target indication system according to the relationships between the moving target and its shadow information in position, dimensions, and intensity. As a sort of feature information available, the moving target shadow may improve the ground target detection performance. A critical precondition for shadow utilization is to obtain the good detection performance for the moving target shadow. However, shadow detection performance will deteriorate inevitably as a result of target motion that blurs its shadow. To address this issue, a knowledge-aided shadow detection algorithm with adaptive thresholds is proposed to improve the shadow detection performance in the developed framework. Furthermore, the theoretical performance analysis is performed, which indicates that the proposed knowledge-aided shadow detection algorithm has a better performance than that of the conventional shadow detection algorithm with a fixed threshold. Finally, numerical simulation experiments are presented to demonstrate that the developed framework can obtain good results for extended ground moving target detection.

Fang Liu;Jie Wu;Lingling Li;Licheng Jiao;Hongxia Hao;Xiangrong Zhang; "A Hybrid Method of SAR Speckle Reduction Based on Geometric-Structural Block and Adaptive Neighborhood," vol.56(2), pp.730-748, Feb. 2018. Given the improvement of synthetic aperture radar (SAR) imaging technologies, the resolution of SAR image is largely improved and the variation of backscatter amplitude should be considered in SAR image processing. In this paper, considering the spatial geometric properties of SAR image in gray pixel space and the sample selection in the estimation of true signal, local directional property of each pixel is explored with the help of SAR sketching method, and two specially designed filters are integrated for adaptive speckle reduction of SAR images. Specifically, based on the sketch map of a SAR image, the orientation of the sketch point lying at each sketch segment is assigned to the corresponding pixel, and thus all pixels of the SAR image are classified as the directional pixels and the nondirectional pixels. For the directional pixels, given the significant directionality of its neighborhood, a geometric-structural block (GB) is built to center on it and GB-wised nonlocal means filter is designed to estimate the true values of all pixels contained in the GB. Moreover, using the local orientation, the whole image is adopted as the searching range to search the similar GBs. For the nondirectional pixels, based on the locally estimated equivalent number of looks, a novel pixel-based metric is proposed to determine the local adaptive neighborhood (AN) with which an AN-based filter is developed to estimate its true value. Besides, since some nondirectional pixels are contained in GBs, a Bayesian-based fusion strategy is designed for the fusion of their estimated values. In the experiments, three synthetic speckled images and five real SAR images [obtained with different resolutions (e.g., 3, 1, and 0.1 m) and different bands (e.g., X-band, C-band, and Ka-band)] are used for evaluation and analysis. Owing to the usage of local spatial geometric property and the combination of two different filters, the proposed method shows a reas- nable performance among the comparison methods, in terms of the speckle reduction and the details’ preservation.

Sen Jia;Bin Deng;Jiasong Zhu;Xiuping Jia;Qingquan Li; "Local Binary Pattern-Based Hyperspectral Image Classification With Superpixel Guidance," vol.56(2), pp.749-759, Feb. 2018. Since it is usually difficult and time-consuming to obtain sufficient training samples by manually labeling, feature extraction, which investigates the characteristics of hyperspectral images (HSIs), such as spectral continuity and spatial locality of surface objects, to achieve the most discriminative feature representation, is very important for HSI classification. Meanwhile, due to the spatial regularity of surface materials, it is desirable to improve the classification performance of HSIs from the superpixel viewpoint. In this paper, we propose a novel local binary pattern (LBP)-based superpixel-level decision fusion method for HSI classification. The proposed framework employs uniform LBP (ULBP) to extract local image features, and then, a support vector machine is utilized to formulate the probability description of each pixel belonging to every class. The composite image of the first three components extracted by a principal component analysis from the HSI data is oversegmented into many homogeneous regions by using the entropy rate segmentation method. Then, a region merging process is applied to make the superpixels obtained more homogeneous and agree with the spatial structure of materials more precisely. Finally, a probability-oriented classification strategy is applied to classify each pixel based on superpixel-level guidance. The proposed framework “ULBP-based superpixel-level decision fusion framework” is named ULBP-SPG. Experimental results on two real HSI data sets have demonstrated that the proposed ULBP-SPG framework is more effective and powerful than several state-of-the-art methods.

Andong Hu;Suqin Wu;Xiaoming Wang;Yan Wang;Robert Norman;Changyong He;Han Cai;Kefei Zhang; "Improvement of Reflection Detection Success Rate of GNSS RO Measurements Using Artificial Neural Network," vol.56(2), pp.760-769, Feb. 2018. Global Navigation Satellite System (GNSS) radio occultation (RO) has been widely used in the prediction of weather, climate, and space weather, particularly in the area of tropospheric analyses. However, one of the issues with GNSS RO measurements is that they are interfered with by the signals reflected from the earth's surface. Many RO events are subject to such interfered GNSS measurements, which are considerably difficult to extract from the GNSS RO measurements. To precisely identify interfered RO events, an improved machine learning approach-a gradient descent artificial neural network (ANN)-aided radio-holography method-is proposed in this paper. Since this method is more complex than most other machine learning methods, for improving its efficiency through the reduction in computational time for near-real-time applications, a scale factor and a regularization factor are also adjusted in the ANN approach. This approach was validated using Constellation Observing System for Meteorology, Ionosphere, and Climate/FC-3 atmPhs (level 1b) data during the period of day of year 172-202, 2015, and its detection results were compared with the flag data set provided by Radio Occultation Meteorology Satellite Application Facilities for the performance assessment and validation of the new approach. The results were also compared with those of the support vector machine method for improvement assessment. The comparison results showed that the proposed method can considerably improve both the success rate of GNSS RO reflection detection and the computational efficiency.

Gencer Sumbul;Ramazan Gokberk Cinbis;Selim Aksoy; "Fine-Grained Object Recognition and Zero-Shot Learning in Remote Sensing Imagery," vol.56(2), pp.770-779, Feb. 2018. Fine-grained object recognition that aims to identify the type of an object among a large number of subcategories is an emerging application with the increasing resolution that exposes new details in image data. Traditional fully supervised algorithms fail to handle this problem where there is low between-class variance and high within-class variance for the classes of interest with small sample sizes. We study an even more extreme scenario named zero-shot learning (ZSL) in which no training example exists for some of the classes. ZSL aims to build a recognition model for new unseen categories by relating them to seen classes that were previously learned. We establish this relation by learning a compatibility function between image features extracted via a convolutional neural network and auxiliary information that describes the semantics of the classes of interest by using training samples from the seen classes. Then, we show how knowledge transfer can be performed for the unseen classes by maximizing this function during inference. We introduce a new data set that contains 40 different types of street trees in 1-ft spatial resolution aerial data, and evaluate the performance of this model with manually annotated attributes, a natural language model, and a scientific taxonomy as auxiliary information. The experiments show that the proposed model achieves 14.3% recognition accuracy for the classes with no training examples, which is significantly better than a random guess accuracy of 6.3% for 16 test classes, and three other ZSL algorithms.

Marie Lachaise;Thomas Fritz;Richard Bamler; "The Dual-Baseline Phase Unwrapping Correction Framework for the TanDEM-X Mission Part 1: Theoretical Description and Algorithms," vol.56(2), pp.780-798, Feb. 2018. The TanDEM-X mission is the first free flying bistatic SAR mission. It has the primary objective to generate within a short time frame a global digital elevation model (DEM) of 10-m absolute vertical accuracy and 2-m relative height accuracy. For that, the whole land mass has been mapped at least twice with different baselines. The success of the mission depends on the accuracy of the final DEM and therefore on the reliability of the phase unwrapping (PU) algorithm. Hence, a robust and versatile PU method, which is in accordance with the acquisition concept, is necessary. This paper presents a new method that combines bistatic high-resolution interferometric data in order to perform an accurate PU on a huge amount of data. The dual-baseline PU correction (DB-PUC) framework addresses this challenge by correcting errors that occurred during the single-baseline PU procedure. It benefits from the additional information available through the differential interferogram and the stereo-radargrammetric phase, which are used to correct region-wise the ambiguity bands of the misestimated unwrapped phases to be less sensitive to noise and possible temporal changes. The multilevel of the DB-PUC approach makes it flexible, computationally efficient, and well adapted to deal with the various PU error scenarios. This framework is used operationally for the processing of the data of the TanDEM-X mission.

Jae-Hyoung Cho;Ji-Hyun Jung;Se-Yun Kim; "Analysis of Cross-Borehole Pulse Radar Signatures on a Terminated Tunnel With Various Penetration Lengths," vol.56(2), pp.799-807, Feb. 2018. A cross-borehole pulse radar system was operated to detect an intrusive man-made tunnel terminated just 1.2 m away from the line of sight between a newly drilled borehole pair at a tunnel test site. Unlike conventional radar signatures on a fully penetrated air-filled tunnel, the relatively fast arrival in the measured time-of-arrival (TOA) profile was highly suppressed at the depth of the terminated tunnel. To analyze the TOA contraction at a terminated tunnel without drilling additional borehole pairs, a finite-difference time-domain (FDTD) simulator is implemented using the accurately measured location information on the terminated tunnel and borehole pair. The relation curves between the time advance in the TOA profile and the penetration length of the terminated tunnel are plotted in the high and low limits of electrical properties of background rock. To verify the accuracy of our FDTD simulated results, the wideband complex permittivity profiles of the core rock samples’ boring at the tunnel test site are measured using an open-ended coaxial probe method. The calculated time advances agree well with the measured values in both cases of fully penetrated and closely terminated borehole pairs in the test site. The presented time advance curves for various penetration lengths will be a valuable guideline on detection of a terminated tunnel in site.

Xinxin Liu;Huanfeng Shen;Qiangqiang Yuan;Xiliang Lu;Chunping Zhou; "A Universal Destriping Framework Combining 1-D and 2-D Variational Optimization Methods," vol.56(2), pp.808-822, Feb. 2018. Striping effects are a common phenomenon in remote-sensing imaging systems, and they can exhibit considerable differences between different sensors. Such artifacts can greatly degrade the quality of the measured data and further limit the subsequent applications in higher level remote-sensing products. Although a lot of destriping methods have been proposed to date, a few of them are robust to different types of stripes. In this paper, we conduct a thorough feature analysis of stripe noise from a novel perspective. With regard to the problem of striping diversity and complexity, we propose a universal destriping framework. In the proposed destriping procedure, a 1-D variational method is first designed and utilized to estimate the statistical feature-based guidance. The guidance information is then incorporated into 2-D optimization to control the image estimation for a reliable and clean output. The iteratively reweighted least-squares method and alternating direction method of multipliers are exploited in the proposed approach to solve the minimization problems. Experiments under various cases of simulated and real stripes confirm the effectiveness and robustness of the proposed model in terms of the qualitative and quantitative comparisons with other approaches.

Marica Baldoncini;Matteo Albéri;Carlo Bottardi;Brian Minty;Kassandra G. C. Raptis;Virginia Strati;Fabio Mantovani; "Airborne Gamma-Ray Spectroscopy for Modeling Cosmic Radiation and Effective Dose in the Lower Atmosphere," vol.56(2), pp.823-834, Feb. 2018. In this paper, we present the results of an ~5-h airborne gamma-ray survey carried out over the Tyrrhenian Sea in which the height range (77-3066) m has been investigated. Gamma-ray spectroscopy measurements have been performed using the AGRS_16L detector, a module of four 4L NaI(Tl) crystals. The experimental setup was mounted on the Radgyro, a prototype aircraft designed for multisensorial acquisitions in the field of proximal remote sensing. By acquiring high-statistics spectra over the sea (i.e., in the absence of signals having geological origin) and by spanning a wide spectrum of altitudes, it has been possible to split the measured count rate into a constant aircraft component and a cosmic component exponentially increasing with increasing height. The monitoring of the count rate having pure cosmic origin in the >3-MeV energy region allowed to infer the background count rates in the 40K, 214Bi, and 208Tl photopeaks, which need to be subtracted in processing airborne gamma-ray data in order to estimate the potassium, uranium, and thorium abundances in the ground. Moreover, a calibration procedure has been carried out by implementing the CARI-6P and Excel-based program for calculating atmospheric cosmic ray spectrum dosimetry tools, according to which the annual cosmic effective dose to human population has been linearly related to the measured cosmic count rates.

Andreas Danklmayer;Jörg Förster;Vincent Fabbro;Gregor Biegel;Thorsten Brehm;Paul Colditz;Laurent Castanet;Yvonick Hurtaud; "Radar Propagation Experiment in the North Sea: The Sylt Campaign," vol.56(2), pp.835-846, Feb. 2018. This paper describes an experiment that was carried out in the North Sea off the Sylt island in May 2012 with the aim to study the influence of the maritime boundary layer conditions on the propagation of radar signals under low grazing angle geometry and to establish a sea clutter database at different frequencies with a view to contribute to new sea clutter models. The radar measurements were carried out with the highly versatile radar called MEMPHIS operating in sea configuration at X-, Ka-, and W-band, simultaneously. As concerns the oceanographic and atmospheric characterization, the collection of measurements was done with a sophisticated suite of sensors partly mounted on the research vessel (RV) Elisabeth Mann Borgese (EMB) and onboard different types of buoys, a catamaran, and a tethered balloon. Over a period of four days, a comprehensive and valuable data set was successfully collected including clutter measurements under different geometrical configurations and propagation runs with corner reflectors mounted onboard RV EMB. An insight into the overall approach is given together with many measurement examples for a very detailed oceanographic and meteorological characterization and a vast number of multifrequency radar acquisitions, showing the complexity of different parameters that have to be considered for sensor performance assessment and prediction.

Zilong Zhong;Jonathan Li;Zhiming Luo;Michael Chapman; "Spectral–Spatial Residual Network for Hyperspectral Image Classification: A 3-D Deep Learning Framework," vol.56(2), pp.847-858, Feb. 2018. In this paper, we designed an end-to-end spectral–spatial residual network (SSRN) that takes raw 3-D cubes as input data without feature engineering for hyperspectral image classification. In this network, the spectral and spatial residual blocks consecutively learn discriminative features from abundant spectral signatures and spatial contexts in hyperspectral imagery (HSI). The proposed SSRN is a supervised deep learning framework that alleviates the declining-accuracy phenomenon of other deep learning models. Specifically, the residual blocks connect every other 3-D convolutional layer through identity mapping, which facilitates the backpropagation of gradients. Furthermore, we impose batch normalization on every convolutional layer to regularize the learning process and improve the classification performance of trained models. Quantitative and qualitative results demonstrate that the SSRN achieved the state-of-the-art HSI classification accuracy in agricultural, rural–urban, and urban data sets: Indian Pines, Kennedy Space Center, and University of Pavia.

Benfeng Wang;Wenkai Lu; "An Efficient Amplitude-Preserving Generalized S Transform and Its Application in Seismic Data Attenuation Compensation," vol.56(2), pp.859-866, Feb. 2018. The time-frequency analysis tools, which are very useful for anomaly identification, reservoir characterization, seismic data processing, and interpretation, are widely used in discrete signal analysis. Among these methods, the generalized S transform (GST) is more flexible, because its analytical window can be self-adjusted according to the local frequency components of the selected discrete signal, besides there exist another two adjustable parameters to make it superior to the S transform (ST). But the amplitude-preserving ability is a little poor near the boundary because the analytical windows do not satisfy the partition of unity, which is a sufficient condition for amplitude-preserving time-frequency transforms. In order to make the GST with the amplitude-preserving ability, we first design a new analytical window, and then propose an amplitude-preserving GST (APGST), but with a higher computational cost. To accelerate the APGST, we provide two strategies: the 3$sigma$ criterion in the probability theory is introduced to accelerate the analytical windows summation and a convolution operator is derived to accelerate the time integral or summation, which generates an efficient APGST (EAPGST). Finally, the proposed EAPGST is used for seismic data attenuation compensation to improve the vertical resolution. Detailed numerical examples are used to demonstrate the validity of the proposed EAPGST in amplitude preserving and high efficiency. Field data attenuation compensation result further proves its successful application in improving the vertical resolution. Besides, the proposed EAPGST can be easily extended into other applications in discrete signal analysis, and remote-sensing and seismology fields.

Xudong Kang;Yufan Huang;Shutao Li;Hui Lin;Jon Atli Benediktsson; "Extended Random Walker for Shadow Detection in Very High Resolution Remote Sensing Images," vol.56(2), pp.867-876, Feb. 2018. The existence of shadows in very high resolution satellite images obstructs image interpretation and the following applications, such as target detection and recognition. Traditional shadow detection methods consider only the pixel-level properties, such as color and intensity of image pixels, and thus, may produce errors around object boundaries. To overcome this problem, a novel shadow detection algorithm based on extended random walker (ERW) is proposed by jointly integrating both shadow property and spatial correlations among adjacent pixels. First, a set of training samples is automatically generated via an improved Otsu-based thresholding method. Then, the support vector machine is applied to obtain an initial detection map, which categorizes all the pixels in the scene into shadow and nonshadow. Finally, the initial detection map is refined with the ERW model, which can simultaneously characterize the shadow property and spatial information in satellite images to further improve shadow detection accuracy. Experiments performed on five real remote sensing images demonstrate the superiority of the proposed method over several state-of-the-art methods in terms of detection accuracy.

Bingyang Liang;Chen Qiu;Feng Han;Chunhui Zhu;Na Liu;Hai Liu;Fubo Liu;Guangyou Fang;Qing Huo Liu; "A New Inversion Method Based on Distorted Born Iterative Method for Grounded Electrical Source Airborne Transient Electromagnetics," vol.56(2), pp.877-887, Feb. 2018. A new iterative inversion algorithm is proposed to reconstruct the electrical conductivity profile in a stratified underground medium for the grounded electrical source airborne transient electromagnetic (GREATEM) system. In forward modeling, we simplify the mathematical expressions of the magnetic fields generated by a finite line source in the layered ground to semianalytical forms in order to save the computation time. The Fréchet derivative is derived for the electromagnetic response at the receivers due to a small perturbation of the conductivity in a certain layer underground. The initial expression of the Fréchet derivative has an expensive triple integral and contains the Bessel function in the integrand. It is simplified by partially eliminating the integration along the source line and deriving the analytical expression for the integration in the vertical direction inside the perturbed layer. In the inverse solution, we use the distorted Born iterative method (DBIM). This is the first time that the DBIM is applied to data measured by the GREATEM system. Besides, the forward and inverse procedures are carried out in the frequency domain and based on the Fréchet derivative of a line source. We demonstrate the validity of our forward model, Fréchet derivative, inverse model, and the precision as well as robustness of the inversion algorithm through numerical computation and comparisons. Finally, we apply the inversion algorithm to the measured data and compare the retrieved conductivity to the actual drilling data.

Shiyang Tang;Chunhui Lin;Yu Zhou;Hing Cheung So;Linrang Zhang;Zheng Liu; "Processing of Long Integration Time Spaceborne SAR Data With Curved Orbit," vol.56(2), pp.888-904, Feb. 2018. Long integration time (LIT) indicates high resolution and/or large scene for spaceborne synthetic aperture radar (SAR) imaging and also means that the effects, brought by curved orbit, cannot be ignored. In this paper, considering the curved orbit caused by the relative motion between an SAR sensor in orbit and targets on a rotating planetary surface, the impacts of the LIT on the imaging results are discussed in detail. The analysis suggests that the cross-coupling phase is two-dimensional (2-D) with spatial variation. Employing the 2-D Taylor series expansion, the 2-D linear relationships between the spatially variant and invariant coefficients are derived, which are exploited to improve the echo formulation. Then, we apply the keystone transform (KT) to process the LIT spaceborne SAR data. Unlike the traditional application of the KT, our two proposed methods, which operate, respectively, in azimuth time and azimuth frequency domains, can greatly remove the spatially variant cross-coupling phase. Moreover, implementation considerations including the curved orbit of LIT spaceborne SAR, applicability of two methods, postprocessing for topography error compensation, and computational load are discussed. Simulation results verify the effectiveness of the developed focusing approaches.

Wanying Song;Ming Li;Peng Zhang;Yan Wu;Xiaofeng Tan;Lin An; "Mixture WG $Gamma$ -MRF Model for PolSAR Image Classification," vol.56(2), pp.905-920, Feb. 2018. The WG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula> model has been validated as an effective model for the characteristic of polarimetric synthetic aperture radar (PolSAR) data statistics. However, due to the complexity of natural scene and the influence of coherent wave, the WG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula> model still needs to be improved to fully consider the polarimetric information. Then, we propose the WG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula> mixture model (WG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>MM) for PolSAR data to maintain the correlations among statistics in PolSAR data. To further consider the spatial-contextual information in PolSAR image classification, we propose a novel mixture model, named mixture WG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>-Markov random field (MWG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>-MRF) model, by introducing the MRF to improve the WG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>MM model for classification. In each law of the MWG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>-MRF model, the interaction term based on the edge penalty function is constructed by the edge-based multilevel-logistic model, while the likelihood term being constructed by the WG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula> model, so that each law of the MWG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>-MRF model can achieve an energy function and has its contribution to the inference of attributive class. Then, the mixture energy function of the MWG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>-MRF model has the fusion of the weig- ted component, given the energy functions of every law. The mixture coefficient and the corresponding mean covariance matrix of the MWG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>-MRF model are estimated by the expectation-maximization algorithm, while the parameters of the WG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula> model being estimated by the method of matrix log-cumulants. Experiments on simulated data and real PolSAR images demonstrate the effectiveness of the MWG<inline-formula> <tex-math notation="LaTeX">$Gamma$ </tex-math></inline-formula>-MRF model and illustrate that it can provide strong noise immunity, get smoother homogeneous areas, and obtain more accurate edge locations.

Thomas D. Neusitzer;Nariman Firoozy;Tyler M. Tiede;Durell S. Desmond;Marcos J. L. Lemes;Gary A. Stern;Søren Rysgaard;Puyan Mojabi;David G. Barber; "Examining the Impact of a Crude Oil Spill on the Permittivity Profile and Normalized Radar Cross Section of Young Sea Ice," vol.56(2), pp.921-936, Feb. 2018. An oil-in-sea ice mesocosm experiment was conducted at the University of Manitoba Sea-Ice Environmental Research Facility from January to March 2016 in which geophysical and electromagnetic parameters of the ice were measured, and general observations about the oil-contaminated ice were made. From the experimental measurements, the presence of crude oil appears to affect the temperature and bulk salinity profiles as well as the normalized radar cross section (NRCS) of the contaminated young sea ice. The measured temperature and bulk salinity profiles of the ice, as well as the crude oil distribution within the ice, were used to model the permittivity profile of the oil-contaminated ice by adapting two mixture models commonly used to describe sea ice to account for the presence of oil. Permittivity modeling results were used to simulate the NRCS of the oil-contaminated sea ice in an effort to determine the accuracy of the models. In addition, the application of X-ray microtomography in modeling the dielectric profile of oil-contaminated sea ice was examined. The sensitivity of the permittivity models for oil-contaminated sea ice to changes in temperature, frequency, and oil volume fraction was also examined.

Xiaodong Xu;Wei Li;Qiong Ran;Qian Du;Lianru Gao;Bing Zhang; "Multisource Remote Sensing Data Classification Based on Convolutional Neural Network," vol.56(2), pp.937-949, Feb. 2018. As a list of remotely sensed data sources is available, how to efficiently exploit useful information from multisource data for better Earth observation becomes an interesting but challenging problem. In this paper, the classification fusion of hyperspectral imagery (HSI) and data from other multiple sensors, such as light detection and ranging (LiDAR) data, is investigated with the state-of-the-art deep learning, named the two-branch convolution neural network (CNN). More specific, a two-tunnel CNN framework is first developed to extract spectral-spatial features from HSI; besides, the CNN with cascade block is designed for feature extraction from LiDAR or high-resolution visual image. In the feature fusion stage, the spatial and spectral features of HSI are first integrated in a dual-tunnel branch, and then combined with other data features extracted from a cascade network. Experimental results based on several multisource data demonstrate the proposed two-branch CNN that can achieve more excellent classification performance than some existing methods.

Yansheng Li;Yongjun Zhang;Xin Huang;Hu Zhu;Jiayi Ma; "Large-Scale Remote Sensing Image Retrieval by Deep Hashing Neural Networks," vol.56(2), pp.950-965, Feb. 2018. As one of the most challenging tasks of remote sensing big data mining, large-scale remote sensing image retrieval has attracted increasing attention from researchers. Existing large-scale remote sensing image retrieval approaches are generally implemented by using hashing learning methods, which take handcrafted features as inputs and map the high-dimensional feature vector to the low-dimensional binary feature vector to reduce feature-searching complexity levels. As a means of applying the merits of deep learning, this paper proposes a novel large-scale remote sensing image retrieval approach based on deep hashing neural networks (DHNNs). More specifically, DHNNs are composed of deep feature learning neural networks and hashing learning neural networks and can be optimized in an end-to-end manner. Rather than requiring to dedicate expertise and effort to the design of feature descriptors, we can automatically learn good feature extraction operations and feature hashing mapping under the supervision of labeled samples. To broaden the application field, DHNNs are evaluated under two representative remote sensing cases: scarce and sufficient labeled samples. To make up for a lack of labeled samples, DHNNs can be trained via transfer learning for the former case. For the latter case, DHNNs can be trained via supervised learning from scratch with the aid of a vast number of labeled samples. Extensive experiments on one public remote sensing image data set with a limited number of labeled samples and on another public data set with plenty of labeled samples show that the proposed remote sensing image retrieval approach based on DHNNs can remarkably outperform state-of-the-art methods under both of the examined conditions.

Sandy Peischl;Jeffrey P. Walker;Dongryeol Ryu;Yann H. Kerr; "Analysis of Data Acquisition Time on Soil Moisture Retrieval From Multiangle L-Band Observations," vol.56(2), pp.966-971, Feb. 2018. This paper investigated the sensitivity of passive microwave L-band soil moisture (SM) retrieval from multiangle airborne brightness temperature data obtained under morning and afternoon conditions from the National Airborne Field Experiment conducted in southeast Australia in 2006. Ground measurements at a dryland focus farm including soil texture, soil temperature, and vegetation water content were used as ancillary data to drive the retrieval model. The derived SM was then in turn evaluated with the ground-measured near-surface SM patterns. The results of this paper show that the Soil Moisture and Ocean Salinity target accuracy of 0.04 m3·m-3 for single-SM retrievals is achievable irrespective of the 6 A.M. and 6 P.M. overpass acquisition times for moisture conditions ≤0.15 m3·m-3. Additional tests on the use of the air temperature as proxy for the vegetation temperature also showed no preference for the acquisition time. The performance of multiparameter retrievals of SM and an additional parameter proved to be satisfactory for SM modeling-independent of the acquisition time-with root-mean-square errors less than 0.06 m3·m-3 for the focus farm.

Hossein Aghababaee;Mahmod Reza Sahebi; "Model-Based Target Scattering Decomposition of Polarimetric SAR Tomography," vol.56(2), pp.972-983, Feb. 2018. When dealing with forest scenario, target scattering separation using synthetic aperture radar (SAR) tomography is a challenging task for the application of biophysical parameter retrieval approaches. One important and widely popular solution used to investigate the scattering mechanism separation based on multipolarimetric multibaseline (MPMB) SAR data is the sum of Kronecker products (SKPs), which provides the basis for decomposition of the data into ground-only and canopy-only contributions. In this paper, we investigate the possibility of characterizing multiple scattering mechanisms using the SKPs of covariance matrix. In particular, we present a method for characterization of forest structure using MPMB data that adapt SKP with the generalized volume description and the physical model of interferometric cross correlation as the sum of scattering contributions. According to the Freeman-Durden model, the method expresses the estimated covariance matrix in terms of the Kronecker product of polarimetric and interferometric coherence matrices corresponding to direct, double-bounce, and random-volume scattering mechanisms. The proposed method is tested with simulated and P-band MB data acquired by ONERA over a tropical forest in French Guiana in the frame of the European Space Agency's campaign TROPISAR. Comparison of the retrieved height of trees with a LiDAR-based canopy model as a reference showed that the proposed method has the potential to decrease root-mean-square error of forest height by up to 3.9 m with respect to SKP.

Gemine Vivone;Rocco Restaino;Jocelyn Chanussot; "A Regression-Based High-Pass Modulation Pansharpening Approach," vol.56(2), pp.984-996, Feb. 2018. Pansharpening usually refers to the fusion of a high spatial resolution panchromatic (PAN) image with a higher spectral resolution but coarser spatial resolution multispectral (MS) image. Owing to the wide applicability of related products, the literature has been populated by many papers proposing several approaches and studies about this issue. Many solutions require a preliminary spectral matching phase wherein the PAN image is matched with the MS bands. In this paper, we propose and properly justify a new approach for performing this step, demonstrating that it yields state-of-the-art performance. The comparison with existing spectral matching procedures is performed by employing four data sets, concerning different kinds of landscapes, acquired by the Pléiades, WorldView-2, and GeoEye-1 sensors.

Yanqing Xie;Yong Xue;Yahui Che;Jie Guang;Linlu Mei;Dave Voorhis;Cheng Fan;Lu She;Hui Xu; "Ensemble of ESA/AATSR Aerosol Optical Depth Products Based on the Likelihood Estimate Method With Uncertainties," vol.56(2), pp.997-1007, Feb. 2018. Within the European Space Agency Climate Change Initiative (CCI) project Aerosol_cci, there are three aerosol optical depth (AOD) data sets of Advanced Along-Track Scanning Radiometer (AATSR) data. These are obtained using the ATSR-2/ATSR dual-view aerosol retrieval algorithm (ADV) by the Finnish Meteorological Institute, the Oxford-Rutherford Appleton Laboratory (RAL) Retrieval of Aerosol and Cloud (ORAC) algorithm by the University of Oxford/RAL, and the Swansea algorithm (SU) by the University of Swansea. The three AOD data sets vary widely. Each has unique characteristics: the spatial coverage of ORAC is greater, but the accuracy of ADV and SU is higher, so none is significantly better than the others, and each has shortcomings that limit the scope of its application. To address this, we propose a method for converging these three products to create a single data set with higher spatial coverage and better accuracy. The fusion algorithm consists of three parts: the first part is to remove the systematic errors; the second part is to calculate the uncertainty and fusion of data sets using the maximum likelihood estimate method; and the third part is to mask outliers with a threshold of 0.12. The ensemble AOD results show that the spatial coverage of fused data set after mask is 148%, 13%, and 181% higher than those of ADV, ORAC, and SU, respectively, and the root-mean-square error, mean absolute error, mean bias error, and relative mean bias are superior to those of the three original data sets. Thus, the accuracy and spatial coverage of the fused AOD data set masked with a threshold of 0.12 are improved compared to the original data set. Finally, we discuss the selection of mask thresholds.

Yong Han;Yong Chen; "Calibration Algorithm for Cross-Track Infrared Sounder Full Spectral Resolution Measurements," vol.56(2), pp.1008-1016, Feb. 2018. The cross-track infrared sounder has been operated in the full spectral resolution (FSR) mode since December 4, 2014. To provide the FSR radiance spectra with a spectral resolution of 0.625 cm−1 for all the three bands, a new calibration algorithm has been developed and implemented for operational uses. The algorithm is an improvement over the previous algorithm that had been operationally used until March 2017. Major changes include the calibration equation, self-apodization correction and resampling matrices, and calibration filter. Compared to the previous algorithm, the improvement reduces the calibration inconsistencies among the nine fields of view and between the forward and reverse interferometer sweep directions by up to 0.5 K, and the differences between observed and simulated spectra by up to 0.4 K.

Roee Diamant;Dror Kipnis;Michele Zorzi; "A Clustering Approach for the Detection of Acoustic/Seismic Signals of Unknown Structure," vol.56(2), pp.1017-1029, Feb. 2018. We focus on the detection of sporadic low-power acoustic/seismic signals of unknown structure and statistics, such as the detection of sound produced by marine mammals, low-power underground signals, or the discovery of events such as volcano eruptions. In these cases, since the ambient noise may be fast time varying and may include many noise transients, threshold-based detection may lead to a significant false alarm rate. Instead, we propose a detection scheme that avoids the use of a decision threshold. Our method is based on clustering the samples of the observed buffer according to a binary hidden Markov model to discriminate between “noise” and “signal” states. Our detector is a modification of the Baum–Welch algorithm that takes into account the expected continuity of the desired signal and obtains a robust detection using the complex but flexible general Gaussian mixture model. The result is a combination of a constrained expectation-maximization algorithm with the Viterbi algorithm. We evaluate the performance of our scheme in numerical simulations, in a seimic test, and in an ocean experiment. The results are close to the hybrid Cramér–Rao lower bound and show that, at the cost of some additional complexity, our proposed algorithm outperforms common benchmark methods in terms of detection and false alarm rates, and also achieves a better accuracy of the time of detection. To allow reproducibility of the results, we publish our code.

Penghui Huang;Xiang-Gen Xia;Guisheng Liao;Zhiwei Yang;Jianjiang Zhou;Xingzhao Liu; "Ground Moving Target Refocusing in SAR Imagery Using Scaled GHAF," vol.56(2), pp.1030-1045, Feb. 2018. In this paper, a new method is proposed to refocus a ground moving target in synthetic aperture radar imagery. In this method, range migration is compensated in the 2-D frequency domain, which can easily be implemented by using the complex multiplications, the fast Fourier transform (FFT), and the inverse FFT operations. Then, the received target signal in a range gate is characterized as a quadratic frequency-modulated (QFM) signal. Finally, a novel parameter estimation method, i.e., scaled generalized high-order ambiguity function (HAF), is proposed to transform the target signal into a signal on 2-D time–frequency plane and realize the 2-D coherent integration, where the peak position accurately determines the second- and third-order parameters of a QFM signal. Compared with our previously proposed generalized Hough-HAF method, the proposed method can obtain a better target focusing performance, since it can eliminate the incoherent operations in both range and azimuth directions. In addition, the proposed method is computationally efficient, since it is free of searching in the whole target focusing procedure. Both simulated and real data processing results are provided to validate the effectiveness of the proposed algorithm.

Redha Touati;Max Mignotte; "An Energy-Based Model Encoding Nonlocal Pairwise Pixel Interactions for Multisensor Change Detection," vol.56(2), pp.1046-1058, Feb. 2018. Image change detection (CD) is a challenging problem, particularly when images come from different sensors. In this paper, we present a novel and reliable CD model, which is first based on the estimation of a robust similarity-feature map generated from a pair of bitemporal heterogeneous remote sensing images. This similarity-feature map, which is supposed to represent the difference between the multitemporal multisensor images, is herein defined, by specifying a set of linear equality constraints, expressed for each pair of pixels existing in the before-and-after satellite images acquired through different modalities. An estimation of this overconstrained problem, also formulated as a nonlocal pairwise energy-based model, is then carried out, in the least square sense, by a fast linear-complexity algorithm based on a multidimensional scaling mapping technique. Finally, the fusion of different binary segmentation results, obtained from this similarity-feature map by different automatic thresholding algorithms, allows us to precisely and automatically classify the changed and unchanged regions. The proposed method is tested on satellite data sets acquired by real heterogeneous sensor, and the results obtained demonstrate the robustness of the proposed model compared with the best existing state-of-the-art multimodal CD methods recently proposed in the literature.

Chuanfa Chen;Yanyan Li;Na Zhao;Changqing Yan; "Robust Interpolation of DEMs From Lidar-Derived Elevation Data," vol.56(2), pp.1059-1068, Feb. 2018. Light detection and ranging (lidar)-derived elevation data are commonly subjected to outliers due to the boundaries of occlusions, physical imperfections of sensors, and surface reflectance. Outliers have a serious negative effect on the accuracy of digital elevation models (DEMs). To decrease the impact of outliers on DEM construction, we propose a robust interpolation algorithm of multiquadric (MQ) based on a regularized least absolute deviation (LAD) technique. The objective function of the proposed method includes a regularization-based smoothing term and an LAD-based fitting term, respectively, used to smooth noisy samples and resist the influence of outliers. To solve the objective function of the proposed method, we develop a simple scheme based on the split-Bregman iteration algorithm. Results from simulated data sets indicate that when sample points are noisy or contaminated by outliers, the proposed method is more accurate than the classical MQ and two recently developed robust algorithms of MQ for surface modeling. Real-world examples of interpolating 1 private and 11 publicly available airborne lidar-derived data sets demonstrate that the proposed method averagely produces better results than two promising interpolation methods including regularized spline with tension (RST) and gridded data-based robust thin plate spline (RTPS). Specifically, the image of RTPS is too smooth to retain terrain details. Although RST can keep subtle terrain features, it is distorted by some misclassified object points (i.e., pseudooutliers). The proposed method obtains a good tradeoff between resisting the effect of outliers and preserving terrain features. Overall, the proposed method can be considered as an alternative for interpolating lidar-derived data sets potentially including outliers.

Qiang Guo;Hongbing Zhang;Feilong Han;Zuoping Shang; "Prestack Seismic Inversion Based on Anisotropic Markov Random Field," vol.56(2), pp.1069-1079, Feb. 2018. Prestack seismic inversion is an ill-posed problem and must be regularized to stabilize the inverted results. In particular, edge-preserving regularization with prior constraints based on Markov random field (MRF) has proved to be an effective technique for reconstructing subsurface models. However, regularized seismic inversion, based on the standard MRF scheme, typically makes use of isotropic MRF neighborhoods, in which the weighting coefficients of the model gradients are equivalent in all directions. Considering real geological conditions, subsurface formations are expected to be laterally continuous and vertically stratified. Therefore, the anisotropic effects caused by model gradients which vary along different directions should not be ignored. In this paper, we proposed a new prestack seismic inversion method based on anisotropic MRF (AMRF). In this method, AMRF coefficients are incorporated into the standard MRF scheme. These coefficients demonstrate directional variations and gradient dependencies, intended to directly correct the errors caused by the anisotropic model gradients on the prior constraints. In particular, we introduced the anisotropic diffusion method to calculate the AMRF coefficients. The proposed inversion method can effectively remove the anisotropic features of the model gradients and significantly improve the inversion results, especially for geologically layered formations. We demonstrated the effectiveness of the inversion method by both 2-D synthetic test and field data example, which presented encouraging results.

Huazhong Ren;Xin Ye;Rongyuan Liu;Jiaji Dong;Qiming Qin; "Improving Land Surface Temperature and Emissivity Retrieval From the Chinese Gaofen-5 Satellite Using a Hybrid Algorithm," vol.56(2), pp.1080-1090, Feb. 2018. Land surface temperature (LST) is a key surface feature parameter. Temperature and emissivity separation (TES) and split-window (SW) algorithms are two typical LST estimation algorithms that have been applied to a variety of sensors to generate LST products. The TES algorithm can synchronously obtain LST and emissivity, but it requires high accuracy for atmospheric correction of the thermal infrared (TIR) data and does not perform well for surfaces with low spectral emissivity contrast. On the contrary, the SW algorithm can retrieve LST without detailed atmospheric data because the linear or nonlinear combination of brightness temperatures in the two adjacent TIR channels can reduce the atmospheric effect; however, this algorithm requires prior accurate pixel emissivity. Combining the two algorithms can improve the accuracy of LST estimation because the emissivity calculated from the TES algorithm can be used in the SW algorithm, and the LST from the SW algorithm can then be applied to the TES algorithm as an initial value to refine emissivity and LST. This paper investigates the aforementioned hybrid algorithm using Chinese Gaofen-5 satellite data, which will provide four-channel data for TIR at 40 m for synchronously retrieving LST and emissivity. The results showed that the hybrid algorithm was less sensitive to instrument noise and atmospheric data error, and can obtain LST and emissivity with an error less than 1 K and 0.015, respectively, which is better than those obtained with the single TES or SW algorithm. Finally, the hybrid algorithm was tested in simulated image and ground-measured data, and obtained accurate results.

David P. Williams; "The Mondrian Detection Algorithm for Sonar Imagery," vol.56(2), pp.1091-1102, Feb. 2018. A new algorithm called the Mondrian detector has been developed for object detection in high-frequency synthetic aperture sonar (SAS) imagery. If a second (low) frequency-band image is available, the algorithm can seamlessly exploit the additional information via an auxiliary prescreener test. This flexible single-band and multiband functionality fills an important capability gap. The algorithm's overall prescreener component limits the number of potential alarms. The main module of the method then searches for areas that pass a subset of pixel-intensity tests. A new set of reliable classification features has also been developed in the process. The overall framework has been kept uncomplicated intentionally in order to facilitate performance estimation, to avoid requiring dedicated training data, and to permit delayed real-time detection at sea on an autonomous underwater vehicle. The promise of the new algorithm is demonstrated on six substantial data sets of real SAS imagery collected at various geographical sites that collectively exhibit a wide range of diverse seafloor characteristics. The results show that-as with Mondrian's art-simplicity can be powerful.

Pablo Morales-Álvarez;Adrián Pérez-Suay;Rafael Molina;Gustau Camps-Valls; "Remote Sensing Image Classification With Large-Scale Gaussian Processes," vol.56(2), pp.1103-1114, Feb. 2018. Current remote sensing image classification problems have to deal with an unprecedented amount of heterogeneous and complex data sources. Upcoming missions will soon provide large data streams that will make land cover/use classification difficult. Machine-learning classifiers can help at this, and many methods are currently available. A popular kernel classifier is the Gaussian process classifier (GPC), since it approaches the classification problem with a solid probabilistic treatment, thus yielding confidence intervals for the predictions as well as very competitive results to the state-of-the-art neural networks and support vector machines. However, its computational cost is prohibitive for large-scale applications, and constitutes the main obstacle precluding wide adoption. This paper tackles this problem by introducing two novel efficient methodologies for GP classification. We first include the standard random Fourier features approximation into GPC, which largely decreases its computational cost and permits large-scale remote sensing image classification. In addition, we propose a model which avoids randomly sampling a number of Fourier frequencies and alternatively learns the optimal ones within a variational Bayes approach. The performance of the proposed methods is illustrated in complex problems of cloud detection from multispectral imagery and infrared sounding data. Excellent empirical results support the proposal in both computational cost and accuracy.

Amit Angal;Xiaoxiong Xiong;Qiaozhen Mu;David R. Doelling;Rajendra Bhatt;Aisheng Wu; "Results From the Deep Convective Clouds-Based Response Versus Scan-Angle Characterization for the MODIS Reflective Solar Bands," vol.56(2), pp.1115-1128, Feb. 2018. The Terra and Aqua Moderate-Resolution Imaging Spectroradiometer (MODIS) scan mirror reflectance is a function of the angle of incidence (AOI) and was characterized prior to launch by the instrument vendor. The relative change of the prelaunch response versus scan angle (RVS) is tracked and linearly scaled on-orbit using observations at two AOIs of 11.2° and 50.2° corresponding to the moon view and solar diffuser, respectively. As the missions continue to operate well beyond their design life of six years, the assumption of linear scaling between the two AOIs is known to be inadequate in accurately characterizing the RVS, particularly at short wavelengths. Consequently, an enhanced approach of supplementing the on-board measurements with response trends from desert pseudoinvariant calibration sites (PICS) was formulated in MODIS Collection 6 (C6). An underlying assumption for the continued effectiveness of this approach is the long-term (multiyear) and short-term (month to month) stability of the PICS. Previous work has shown that the deep convective clouds (DCC) can also be used to monitor the on-orbit RVS performance with less trend uncertainties compared with desert sites. In this paper, the raw sensor response to the DCC is used to characterize the on-orbit RVS on a band and mirror-side basis. These DCC-based RVS results are compared with those of C6 PICS-based RVS, showing an agreement within 2% observed in most cases. The pros and cons of using a DCC-based RVS approach are also discussed in this paper. Although this reaffirms the efficacy of the C6 PICS-based RVS, the DCC-based RVS approach presents itself as an effective alternative for future considerations. Potential applications of this approach to other instruments, such as Suomi National Polar-orbiting Partnership, Joint Polar Satellite Systems, and Visible Infrared Imaging Radiometer Suite, are also discussed.

Massimo Zanetti;Lorenzo Bruzzone; "A Theoretical Framework for Change Detection Based on a Compound Multiclass Statistical Model of the Difference Image," vol.56(2), pp.1129-1143, Feb. 2018. The change detection (CD) problem is very important in the remote sensing domain. The advent of a new generation of multispectral (MS) sensors has given rise to new challenges in the development of automatic CD techniques. In particular, typical approaches to CD are not able to well model and properly exploit the increased radiometric resolution characterizing new data as this results in a higher sensitivity to the number of natural classes that can be statistically modeled in the images. In this paper, we introduce a theoretical framework for the description of the statistical distribution of the difference image as a compound model where each class is determined by temporally correlated class transitions in the bitemporal images. The potential of the proposed framework is demonstrated on the very common problem of binary CD based on setting a threshold on the magnitude of the difference image. Here, under some simplifying assumptions, a multiclass distribution of the magnitude feature is derived and an unsupervised method based on the expectation–maximization algorithm and Bayes decision is proposed. Its effectiveness is demonstrated on a large variety of data sets from different MS sensors. In particular, experimental tests confirm that: 1) the fitting of the magnitude distribution significantly improves if compared with already existing models and 2) the overall CD error is close to the optimal value.

Bindita Chaudhuri;Begüm Demir;Subhasis Chaudhuri;Lorenzo Bruzzone; "Multilabel Remote Sensing Image Retrieval Using a Semisupervised Graph-Theoretic Method," vol.56(2), pp.1144-1158, Feb. 2018. Conventional supervised content-based remote sensing (RS) image retrieval systems require a large number of already annotated images to train a classifier for obtaining high retrieval accuracy. Most systems assume that each training image is annotated by a single label associated to the most significant semantic content of the image. However, this assumption does not fit well with the complexity of RS images, where an image might have multiple land-cover classes (i.e., multilabels). Moreover, annotating images with multilabels is costly and time consuming. To address these issues, in this paper, we introduce a semisupervised graph-theoretic method in the framework of multilabel RS image retrieval problems. The proposed method is based on four main steps. The first step segments each image in the archive and extracts the features of each region. The second step constructs an image neighborhood graph and uses a correlated label propagation algorithm to automatically assign a set of labels to each image in the archive by exploiting only a small number of training images annotated with multilabels. The third step associates class labels with image regions by a novel region labeling strategy, whereas the final step retrieves the images similar to a given query image by a subgraph matching strategy. Experiments carried out on an archive of aerial images show the effectiveness of the proposed method when compared with the state-of-the-art RS content-based image retrieval methods.

María Díaz;Raúl Guerra;Sebastián López;Roberto Sarmiento; "An Algorithm for an Accurate Detection of Anomalies in Hyperspectral Images With a Low Computational Complexity," vol.56(2), pp.1159-1176, Feb. 2018. Anomaly detection (AD) is an important technique in hyperspectral data analysis that permits to distinguish rare objects with unknown spectral signatures that are particularly not abundant in a scene. In this paper, a novel algorithm for an accurate detection of anomalies in hyperspectral images with a low computational complexity, named ADALOC2, is proposed. It is based on two main processing stages. First, a set of characteristic pixels that best represent both anomaly and background classes are extracted applying orthogonal projection techniques. Second, the abundance maps associated to these pixels are estimated. Under the assumption that the anomaly class is composed of a scarce group of image pixels, rare targets can be identified from abundance maps characterized by a representation coefficient matrix with a large amount of almost zero elements. Unlike the other algorithms of the state of the art, the ADALOC2 algorithm has been specially designed for being efficiently implemented into parallel hardware devices for applications under real-time constraints. To achieve this, the ADALOC2 algorithm uses simple and highly parallelized operations, avoiding to perform complex matrix operations such as the computation of an inverse matrix or the extraction of eigenvalues and eigenvectors. An extensive set of simulations using the most representative state-of-the-art AD algorithms and both real and synthetic hyperspectral data sets have been conducted. Moreover, extra assessment metrics apart from classical receiver operating characteristic curves have been defined in order to make deeper comparisons. The obtained results clearly support the benefits of our proposal, both in terms of the accuracy of the detection results and the processing power demanded.

L. Hettak;H. Roussel;M. Casaletti;R. Mittra; "A Numerically Efficient Method for Predicting the Scattering Characteristics of a Complex Metallic Target Located Inside a Large Forested Area," vol.56(2), pp.1177-1185, Feb. 2018. An efficient electromagnetic model for the scattering analysis of targets placed in large natural environments is presented. A hybrid formulation combining volume and surface integral equations is used to describe forest environment (dielectric objects) in which metallic structures are present. A large part of the forest can be analyzed electromagnetically by using the characteristic basis function method, whose use enables us to simulate the problem at hand and significantly reduce the dimension of the linear system that needs to be solved.

Jun Li;Fei Hu;Feng He;Liang Wu; "High-Resolution RFI Localization Using Covariance Matrix Augmentation in Synthetic Aperture Interferometric Radiometry," vol.56(2), pp.1186-1198, Feb. 2018. Radio frequency interference (RFI) is a significant limiting factor in the retrieval of geophysical parameters measured by microwave radiometers. RFI localization is crucial to mitigate or remove the RFI impacts. In this paper, a novel RFI localization approach using covariance matrix augmentation in synthetic aperture interferometric radiometry (SAIR) is proposed. It utilizes the property of the sparse array configuration, which is commonly used in SAIR, where the sparse array can be viewed as a virtual filled array with much larger number of antenna elements. The approach can be applied in SAIR with a sparse array configuration, such as the European Space Agency Soil Moisture and Ocean Salinity (SMOS) mission. Results on real SMOS data show that, compared with the previous approach, the presented approach has an improved performance of RFI localization with comparable accuracy of localization, such as improved spatial resolution, lower sidelobes, and larger identifiable number of RFIs.

Minh-Tan Pham;Sébastien Lefèvre;Erchan Aptoula; "Local Feature-Based Attribute Profiles for Optical Remote Sensing Image Classification," vol.56(2), pp.1199-1212, Feb. 2018. This paper introduces an extension of morphological attribute profiles (APs) by extracting their local features. The so-called local feature-based APs (LFAPs) are expected to provide a better characterization of each APs’ filtered pixel (i.e., APs’ sample) within its neighborhood, and hence better deal with local texture information from the image content. In this paper, LFAPs are constructed by extracting some simple first-order statistical features of the local patch around each APs’ sample such as mean, standard deviation, and range. Then, the final feature vector characterizing each image pixel is formed by combining all local features extracted from APs of that pixel. In addition, since the self-dual APs (SDAPs) have been proved to outperform the APs in recent years, a similar process will be applied to form the local feature-based SDAPs (LFSDAPs). In order to evaluate the effectiveness of LFAPs and LFSDAPs, supervised classification using both the random forest and the support vector machine classifiers is performed on the very high resolution Reykjavik image as well as the hyperspectral Pavia University data. Experimental results show that LFAPs (respectively, LFSDAPs) can considerably improve the classification accuracy of the standard APs (respectively, SDAPs) and the recently proposed histogram-based APs.

Jihao Yin;Hui Qv;Xiaoyan Luo;Xiuping Jia; "Corrections to “Segment-Oriented Depiction and Analysis for Hyperspectral Image Data” [Jul 17 3982-3996[Name:_blank]]," vol.56(2), pp.1213-1213, Feb. 2018. In [1], information regarding the corresponding author is missing. The information is updated here. The updated footnote below shows that Xiaoyan Luo is the corresponding author for this paper.

Lichao Mou;Pedram Ghamisi;Xiao Xiang Zhu; "Corrections to “Deep Recurrent Neural Networks for Hyperspectral Image Classification” [Jul 17 3639-3655[Name:_blank]]," vol.56(2), pp.1214-1215, Feb. 2018. Here, we correct some errors caused by a programming bug (a data type error) in overall accuracies (OAs) reported in [1]. The corrected OAs are underlined and shown in bold in Tables IIII.

* "IEEE Transactions on Geoscience and Remote Sensing information for authors," vol.56(2), pp.C3-C3, Feb. 2018.* Provides instructions and guidelines to prospective authors who wish to submit manuscripts.

* "IEEE Transactions on Geoscience and Remote Sensing institutional listings," vol.56(2), pp.C4-C4, Feb. 2018.* Advertisements.

IEEE Geoscience and Remote Sensing Letters - new TOC (2018 February 15) [Website]

* "[Front cover/Table of contents[Name:_blank]]," vol.15(2), pp.C1-C1, Feb. 2018.* Presents the cover/table of contents for this issue of the periodical.

* "IEEE Geoscience and Remote Sensing Letters publication information," vol.15(2), pp.C2-C2, Feb. 2018.* Provides a listing of current staff, committee members and society officers.

* "Table of contents," vol.15(2), pp.161-162, Feb. 2018.* Presents the table of contents for this issue of the periodical.

Ken B. Cooper;Raquel Rodriguez Monje;Luis Millán;Matthew Lebsock;Simone Tanelli;Jose V. Siles;Choonsup Lee;Andrew Brown; "Atmospheric Humidity Sounding Using Differential Absorption Radar Near 183 GHz," vol.15(2), pp.163-167, Feb. 2018. A tunable G-band frequency-modulated continuous-wave radar system has been developed and used to perform differential absorption atmospheric humidity measurements for the first time. The radar's transmitter uses high- power-handling GaAs Schottky diodes to generate between 15-23 dBm over a 10-GHz bandwidth near 183 GHz. By virtue of a high-isolation circular polarization duplexer, the monostatic radar's receiver maintains a noise figure of about 7 dB even while the transmitter is on. With an antenna gain of 40 dB, high-SNR detection of light rain is achieved out to several hundred meters distance. Owing to the strong spectral dependence of the atmospheric absorption over the upper flank of the 183-GHz water absorption line, range-resolved measurements of absolute humidity can be obtained by ratioing the rain echoes over both range and frequency. Absorption measurements obtained are consistent with models of atmospheric millimeter-wave attenuation, and they demonstrate a new method for improving the accuracy of humidity measurements inside of clouds.

Bo Du;Yujia Sun;Shihan Cai;Chen Wu;Qian Du; "Object Tracking in Satellite Videos by Fusing the Kernel Correlation Filter and the Three-Frame-Difference Algorithm," vol.15(2), pp.168-172, Feb. 2018. Object tracking is a popular topic in the field of computer vision. The detailed spatial information provided by a very high resolution remote sensing sensor makes it possible to track targets of interest in satellite videos. In recent years, correlation filters have yielded promising results. However, in terms of dealing with object tracking in satellite videos, the kernel correlation filter (KCF) tracker achieves poor results due to the fact that the size of each target is too small compared with the entire image, and the target and the background are very similar. Therefore, in this letter, we propose a new object tracking method for satellite videos by fusing the KCF tracker and a three-frame-difference algorithm. A specific strategy is proposed herein for taking advantage of the KCF tracker and the three-frame-difference algorithm to build a strong tracker. We evaluate the proposed method in three satellite videos and show its superiority to other state-of-the-art tracking methods.

Kaiqiang Chen;Kun Fu;Menglong Yan;Xin Gao;Xian Sun;Xin Wei; "Semantic Segmentation of Aerial Images With Shuffling Convolutional Neural Networks," vol.15(2), pp.173-177, Feb. 2018. Semantic segmentation of aerial images refers to assigning one land cover category to each pixel. This is a challenging task due to the great differences in the appearances of ground objects. Many attempts have been made during the past decades. In recent years, convolutional neural networks (CNNs) have been introduced in the remote sensing field, and various solutions have been proposed to realize dense semantic labeling with CNNs. In this letter, we propose shuffling CNNs to realize semantic segmentation of aerial images in a periodic shuffling manner. This approach is a supplement to current methods for semantic segmentation of aerial images. We propose a naive version and a deeper version of this method, and both are adept at detecting small objects. Additionally, we propose a method called field-of-view (FoV) enhancement that can enhance the predictions. This method can be applied to various networks, and our experiments verify its effectiveness. The final results are further improved through an ensemble method that averages the score maps generated by the models at different checkpoints of the same network. We evaluate our models using the ISPRS Vaihingen and Potsdam data sets, and we acquire promising results using these two data sets.

Ziqiang Ma;Lianqing Zhou;Wu Yu;Yuanyuan Yang;Hongfeng Teng;Z. Shi; "Improving TMPA 3B43 V7 Data Sets Using Land-Surface Characteristics and Ground Observations on the Qinghai–Tibet Plateau," vol.15(2), pp.178-182, Feb. 2018. The accurate knowledge of precipitation information over the Qinghai-Tibet Plateau, where the rain gauge networks are limited, is vital for various applications. While satellite-based precipitation estimates provide high spatial resolution (0.25°), large uncertainties and systematic anomalies still exist over this critical area. To derive more accurate monthly precipitation estimates, a spatial data-mining algorithm was used to remove the obvious anomalies compared with their neighbors from the original Tropical Rainfall Measuring Mission (TRMM) multisatellite precipitation analysis (TMPA) 3B43 V7 data at an annual scale, as the TMPA data are more accurate than other satellite-based precipitation estimates. To supplement the international exchange stations, additional ground observations were used to calibrate and improve the TMPA data with anomalies removed at an annual scale. Finally, a disaggregation strategy was adopted to derive monthly precipitation estimates based on the calibrated TMPA data. We concluded that: 1) the obvious anomalies compared with their neighbors could be removed from the original TMPA data sets and 2) the calibrated results were of a higher quality than the original TMPA data in each month from 2000 to 2013. The improved TMPA 3B43 V7 data sets over the Qinghai-Tibet plateau, named NITMPA3B43_QTP, are available at http://agri.zju.edu.cn/NITMPA3B43_QTP/.

Yishu Liu;Yingbin Liu;Liwang Ding; "Scene Classification Based on Two-Stage Deep Feature Fusion," vol.15(2), pp.183-186, Feb. 2018. In convolutional neural networks (CNNs), higher layer information is more abstract and more task specific, so people usually concern themselves with fully connected (FC) layer features, believing that lower layer features are less discriminative. However, a few researchers showed that the lower layers also provide very rich and powerful information for image representation. In view of these study findings, in this letter, we attempt to adaptively and explicitly combine the activations from intermediate and FC layers to generate a new CNN with directed acyclic graph topology, which is called the converted CNN. After that, two converted CNNs are integrated together to further improve the classification performance. We validate our proposed two-stage deep feature fusion model over two publicly available remote sensing data sets, and achieve a state-of-the-art performance in scene classification tasks.

Jérémy E. Cohen;Nicolas Gillis; "Spectral Unmixing With Multiple Dictionaries," vol.15(2), pp.187-191, Feb. 2018. Spectral unmixing aims at recovering the spectral signatures of materials, called endmembers, mixed in a hyperspectral image (HSI) or multispectral image, along with their abundances. A typical assumption is that the image contains one pure pixel per endmember, in which case spectral unmixing reduces to identifying these pixels. Many fully automated methods have been proposed in recent years, but little work has been done to allow users to select areas where pure pixels are present manually or using a segmentation algorithm. Additionally, in a nonblind approach, several spectral libraries may be available rather than a single one, with a fixed number (or an upper or lower bound) of endmembers to chose from each. In this letter, we propose a multiple-dictionary constrained low-rank matrix approximation model that addresses these two problems. We propose an algorithm to compute this model, dubbed multiple matching pursuit alternating least squares, and its performance is discussed on both synthetic and real HSIs.

Antonio Maria Garcia Tommaselli;Maurício Galo;Thiago Tiedtke dos Reis;Roberto da Silva Ruy;Marcus Vinicius Antunes de Moraes;Wander Vieira Matricardi; "Development and Assessment of a Data Set Containing Frame Images and Dense Airborne Laser Scanning Point Clouds," vol.15(2), pp.192-196, Feb. 2018. This letter describes the main features of a data set that contains aerial images acquired with a medium format digital camera and point clouds collected using an airborne laser scanning unit, as well as ground control points and direct georeferencing data. The flights were performed in 2014 over an urban area in Presidente Prudente, São Paulo, Brazil, using different flight heights. These flights covered several features of interest for research, including buildings of different sizes and roof materials, roads, and vegetation. Three point clouds with different densities, a block of digital aerial images, and auxiliary data are available. A geometric assessment was conducted to ensure the accuracy and consistency of the data, and an RMSE of 7 cm was achieved using bundle block adjustment. The data set is freely available for download, and it will be expanded with data collected over time.

Haofeng Dou;Qingxia Li;Liangqi Gui;Ke Chen;Yufang Li;Congcong Huang;Menglin Hu; "One-Dimensional Mirrored Aperture Synthesis With Rotating Reflector," vol.15(2), pp.197-201, Feb. 2018. In this letter, 1-D mirrored aperture synthesis with a rotating reflector (1-D MAS-R) is proposed to improve the spatial resolution and reduce the number of required antennas for passive microwave remote sensing. The principle of the 1-D MAS-R with an antenna array is given, and from the principle, the 1-D MAS-R with only one antenna can also reconstruct the image of the scene. Simulation results demonstrate the validity of the 1-D MAS-R even with only one antenna, and the spatial resolution is improved by increasing the distance between the reflector and the antenna or antenna array.

Panagiotis Sismanidis;Benjamin Bechtel;Iphigenia Keramitsoglou;Chris T. Kiranoudis; "Mapping the Spatiotemporal Dynamics of Europe’s Land Surface Temperatures," vol.15(2), pp.202-206, Feb. 2018. The land surface temperature (LST) drives many terrestrial biophysical processes and varies rapidly in space and time primarily due to the earth's diurnal and annual cycles. Models of the diurnal and annual LST cycle retrieved from satellite data can be reduced to several gap-free parameters that represent the surface's thermal characteristics and provide a generalized characterization of the LST temporal dynamics. In this letter, we use such an approach to map Europe's annual and diurnal LST dynamics. In particular, we reduce a five-year time series (2009-2013) of diurnal LST from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) to 48 sets of half-hourly annual cycle parameters (ACPs), namely, the mean annual LST, the yearly amplitude of LST, and the LST phase shift from the spring equinox. The derived data provide a complete representation of how mainland Europe responds to the heating of the sun and the nighttime LST decay and reveal how Europe's biogeographic regions differ in that respect. We further argue that the SEVIRI ACP can provide an observation-based spatially consistent background for studying and characterizing the thermal behavior of the surface and also a data set to support climate classification at a finer spatial resolution.

Yuting Yang;Junyu Dong;Xin Sun;Estanislau Lima;Quanquan Mu;Xinhua Wang; "A CFCC-LSTM Model for Sea Surface Temperature Prediction," vol.15(2), pp.207-211, Feb. 2018. Sea surface temperature (SST) prediction is not only theoretically important but also has a number of practical applications across a variety of ocean-related fields. Although a large amount of SST data obtained via remote sensor are available, previous work rarely attempted to predict future SST values from history data in spatiotemporal perspective. This letter regards SST prediction as a sequence prediction problem and builds an end-to-end trainable long short term memory (LSTM) neural network model. LSTM naturally has the ability to learn the temporal relationship of time series data. Besides temporal information, spatial information is also included in our LSTM model. The local correlation and global coherence of each pixel can be expressed and retained by patches with fixed dimensions. The proposed model essentially combines the temporal and spatial information to predict future SST values. Its structure includes one fully connected LSTM layer and one convolution layer. Experimental results on two data sets, i.e., one Advanced Very High Resolution Radiometer SST data set covering China Coastal waters and one National Oceanic and Atmospheric Administration High-Resolution SST data set covering the Bohai Sea, confirmed the effectiveness of the proposed model.

Ying Zhan;Dan Hu;Yuntao Wang;Xianchuan Yu; "Semisupervised Hyperspectral Image Classification Based on Generative Adversarial Networks," vol.15(2), pp.212-216, Feb. 2018. Because the collection of ground-truth labels is difficult, expensive, and time-consuming, classifying hyperspectral images (HSIs) with few training samples is a challenging problem. In this letter, we propose a novel semisupervised algorithm for the classification of hyperspectral data by training a customized generative adversarial network (GAN) for hyperspectral data. The GAN constructs an adversarial game between a discriminator and a generator. The generator generates samples that are not distinguishable by the discriminator, and the discriminator determines whether or not a sample is composed of real data. We design a semisupervised framework for HSI data based on a 1-D GAN (HSGAN). This framework enables the automatic extraction of spectral features for HSI classification. When HSGAN is trained using unlabeled hyperspectral data, the generator can generate hyperspectral samples that are similar to the real data, while the discriminator contains the features, which can be used to classify hyperspectral data with only a small number of labeled samples. The performance of the HSGAN is evaluated on the Airborne Visible Infrared Imaging Spectrometer image data, and the results show that the proposed framework achieves very promising results with a small number of labeled samples.

Wenlong Niu;Wei Zheng;Zhen Yang;Yong Wu;Balazs Vagvolgyi;Bo Liu; "Moving Point Target Detection Based on Higher Order Statistics in Very Low SNR," vol.15(2), pp.217-221, Feb. 2018. This letter presents an approach for the detection of moving point targets on high-frame-rate image sequences with low spatial resolution and low SNR based on higher order statistical theory. We propose a novel method for analyzing the time-domain evolution of image data for distinguishing between the background and the target in situations when the spatial signal of the target is swamped by noise. Our method is formulated to detect a time-domain transient signal of unknown scale and arrival time in noisy background. We proposed a bispectrum-based model to characterize the temporal behavior of pixels, and the detection ability under different frame rates and SNRs is analyzed. The method is evaluated using both simulated and real-world data, and we provide a comparison to other widely used point target detection approaches. Our experimental results demonstrate that our algorithm can efficiently detect extremely low SNR targets that are virtually invisible to humans based on time-domain analysis of image sequences.

Massimiliano Pieraccini; "Noise Performance Comparison Between Continuous Wave and Stroboscopic Pulse Ground Penetrating Radar," vol.15(2), pp.222-226, Feb. 2018. Although stroboscopic pulse (SP) ground penetrating radar (GPR) is the most popular and widespread equipment for subsoil investigation, continuous-wave (CW) radar has better performance in terms of noise, system dynamic range, and penetration depth, at the expense of greater complexity and cost of the components. The aim of this letter is a direct comparison between SP GPR and CW GPR through an extensive measurement campaign in five different locations representative of the different conditions where a GPR could operate.

Beom-Seok Oh;Xin Guo;Fangyuan Wan;Kar-Ann Toh;Zhiping Lin; "Micro-Doppler Mini-UAV Classification Using Empirical-Mode Decomposition Features," vol.15(2), pp.227-231, Feb. 2018. In this letter, we propose an empirical-mode decomposition (EMD)-based method for automatic multicategory mini-unmanned aerial vehicle (UAV) classification. The radar echo signal is first decomposed into a set of oscillating waveforms by EMD. Then, eight statistical and geometrical features are extracted from the oscillating waveforms to capture the phenomenon of blade flashes. After feature normalization and fusion, a nonlinear support vector machine is trained for target class-label prediction. Our empirical results on real measurement of radar signals show encouraging mini-UAV classification accuracy performance.

Famao Ye;Yanfei Su;Hui Xiao;Xuqing Zhao;Weidong Min; "Remote Sensing Image Registration Using Convolutional Neural Network Features," vol.15(2), pp.232-236, Feb. 2018. Successful remote sensing image registration is an important step for many remote sensing applications. The scale-invariant feature transform (SIFT) is a well-known method for remote sensing image registration, with many variants of SIFT proposed. However, it only uses local low-level information, and loses much middle- or high-level information to register. Image features extracted by a convolutional neural network (CNN) have achieved the state-of-the-art performance for image classification and retrieval problems, and can provide much middle- and high-level information for remote sensing image registration. Hence, in this letter, we investigate how to calculate the CNN feature, and study the way to fuse SIFT and CNN features for remote sensing image registration. The experimental results demonstrate that the proposed method yields a better registration performance in terms of both the aligning accuracy and the number of correct correspondences.

Qingyun Yan;Weimin Huang;Giuseppe Foti; "Quantification of the Relationship Between Sea Surface Roughness and the Size of the Glistening Zone for GNSS-R," vol.15(2), pp.237-241, Feb. 2018. A formulation of the relationship between sea-surface roughness and extension of the glistening zone (GZ) of a Global Navigation Satellite System Reflectometry (GNSS-R) system is presented. First, an analytical expression of the link between GZ area, viewing geometry, and surface mean square slope (MSS) is derived. Then, a strategy for retrieval of surface roughness from the delay-Doppler map (DDM) is illustrated, including details of data preprocessing, quality control, and GZ area estimation from the DDM. Next, an example for application of the proposed approach to spaceborne GNSS-R remote sensing is provided, using DDMs from the TechDemoSat-1 mission. The algorithm is first calibrated using collocated in situ roughness estimates using data sets from the National Data Buoy Center, its retrieval performance is then assessed, and some of the limitations of the suggested technique are discussed. Overall, good correlation is found between buoy-derived MSS and estimates obtained using the proposed strategy ($r=0.73$ ).

Yue Wu;Qiguang Miao;Wenping Ma;Maoguo Gong;Shanfeng Wang; "PSOSAC: Particle Swarm Optimization Sample Consensus Algorithm for Remote Sensing Image Registration," vol.15(2), pp.242-246, Feb. 2018. Image registration is an important preprocessing step for many remote sensing image processing applications, and its result will affect the performance of the follow-up procedures. Establishing reliable matches is a key issue in point matching-based image registration. Due to the significant intensity mapping difference between remote sensing images, it may be difficult to find enough correct matches from the tentative matches. In this letter, particle swarm optimization (PSO) sample consensus algorithm is proposed for remote sensing image registration. Different from random sample consensus (RANSAC) algorithm, the proposed method directly samples the modal transformation parameter rather than randomly selecting tentative matches. Thus, the proposed method is less sensitive to the correct rate than RANSAC, and it has the ability to handle lower correct rate and more matches. Meanwhile, PSO is utilized to optimize parameter as its efficiency. The proposed method is tested on several multisensor remote sensing image pairs. The experimental results indicate that the proposed method yields a better registration performance in terms of both the number of correct matches and aligning accuracy.

Jingtian Tang;Shuanggui Hu;Zhengyong Ren;Chaojian Chen; "Localization of Multiple Underwater Objects With Gravity Field and Gravity Gradient Tensor," vol.15(2), pp.247-251, Feb. 2018. We present a novel algorithm to locate multiple underwater objects in real time using gravity field vector and gravity gradient tensor signals. This algorithm formulates the task of localization of multiple underwater objects into a regularized nonlinear problem, which is solved with the standard Levenberg–Marquardt algorithm. The regularization parameters are estimated by cross validation. The initial coordinates and masses of these underwater objects are automatically determined by solving a single-object localization problem. A synthetic navigation model with two underwater objects was adopted to validate the proposed algorithm. The results show that it has good stability and antinoise ability for multiple underwater objects localizations.

Andrea Carolina Flores Rodriguez;Gustavo Fraidenraich;Tarcísio A. P. Soares;José Cândido S. Santos Filho;Marco A. M. Miranda;Michel Daoud Yacoub; "Optimal and Suboptimal Velocity Estimators for ArcSAR With Distributed Target," vol.15(2), pp.252-256, Feb. 2018. Two new methods of radial velocity estimation for distributed targets in arc-scanning synthetic aperture radar (ArcSAR) systems, namely, the maximum-likelihood estimator (MLE) and the suboptimal method based on the least squares estimation (LSE), are proposed, derived, and analyzed. To this end, we establish that <inline-formula> <tex-math notation="LaTeX">$n$ </tex-math></inline-formula> scatterers of the distributed target are uniformly dispersed within the radar resolution cell of dimensions <inline-formula> <tex-math notation="LaTeX">$a times b$ </tex-math></inline-formula> and they move randomly at different velocities. Furthermore, the effect of the antenna pattern is considered to characterize the amplitude of the scattered signal. Thus, from the coherent integration of the scatters at each pulse repetition interval in radar scanning, <inline-formula> <tex-math notation="LaTeX">$m$ </tex-math></inline-formula> data sequences are obtained as samples of the composite signal, which follows a multivariate normal distribution. From this, the covariance matrix, upon which the methods are based, is derived. Simulations have been carried out to compare the new methods with existing methods, namely, phase, energy, and correlation, as a function of the signal-to-noise ratio. Finally, the results show that the MLE and LSE methods outperform the conventional methods, providing a gain of more than 10 dB.

Naisen Yang;Hong Tang;Hongquan Sun;Xin Yang; "DropBand: A Simple and Effective Method for Promoting the Scene Classification Accuracy of Convolutional Neural Networks for VHR Remote Sensing Imagery," vol.15(2), pp.257-261, Feb. 2018. The dropout and data augmentation techniques are widely used to prevent a convolutional neural network (CNN) from overfitting. However, the dropout technique does not work well when applied to the input channels of neural networks, and data augmentation is usually employed along the image plane. In this letter, we present DropBand, which is a simple and effective method of promoting the classification accuracy of CNNs for very-high-resolution remote sensing image scenes. In DropBand, more training samples are generated by dropping certain spectral bands out of original images. Furthermore, all samples with the same set of spectral bands are collected together to train a base CNN. The final prediction for a test sample is represented by the combination of outputs of all base CNNs. The experimental results for three publicly available data sets, i.e., the SAT-4, SAT-6, and UC-Merced image data sets, show that DropBand can significantly improve the classification accuracy of a CNN.

P. V. Arun;K. M. Buddhiraju;A. Porwal; "Integration of Contextual Knowledge in Unsupervised Subpixel Classification: Semivariogram and Pixel-Affinity Based Approaches," vol.15(2), pp.262-266, Feb. 2018. This letter investigates the use of coarse-image features for predicting class labels at a given finer spatial scale. In this regard, two unsupervised subpixel mapping approaches, a semivariogram method, and a pixel-affinity based method are proposed. Furthermore, segmentation-based spectral unmixing is explored so as to address the spectral variability and nonconvexity of classes. In addition, the gradient information is employed to resolve uncertainties in the unmixing process. The proposed modifications based on pixel-affinity and semivariogram have produced an accuracy improvement of 5% or more over the state-of-the-art approaches.

Chunsun Zhang;Yuxiang He;Clive S. Fraser; "Spectral Clustering of Straight-Line Segments for Roof Plane Extraction From Airborne LiDAR Point Clouds," vol.15(2), pp.267-271, Feb. 2018. This letter presents a novel approach to automated extraction of roof planes from airborne light detection and ranging data based on spectral clustering of straight-line segments. The straight-line segments are derived from laser scan lines, and 3-D line geometry analysis is employed to identify coplanar line segments so as to avoid skew lines in plane estimation. Spectral analysis reveals the spectrum of the adjacency matrix formed by the straight-line segments. Spectral clustering is then performed in feature space where the clusters are more prominent, resulting in a more robust extraction of roof planes. The proposed approach has been tested on ISPRS benchmark data sets, with the results showing high quality in terms of completeness, correctness, and geometrical accuracy, thus confirming that the proposed approach can extract roof planes both accurately and efficiently.

Sanyi Yuan;Jiwei Liu;Shangxu Wang;Tieyi Wang;Peidong Shi; "Seismic Waveform Classification and First-Break Picking Using Convolution Neural Networks," vol.15(2), pp.272-276, Feb. 2018. Regardless of successful applications of the convolutional neural networks (CNNs) in different fields, its application to seismic waveform classification and first-break (FB) picking has not been explored yet. This letter investigates the application of CNNs for classifying time-space waveforms from seismic shot gathers and picking FBs of both direct wave and refracted wave. We use representative subimage samples with two types of labeled waveform classification to supervise CNNs training. The goal is to obtain the optimal weights and biases in CNNs, which are solved by minimizing the error between predicted and target label classification. The trained CNNs can be utilized to automatically extract a set of time-space attributes or features from any subimage in shot gathers. These attributes are subsequently inputted to the trained fully connected layer of CNNs to output two values between 0 and 1. Based on the two-element outputs, a discriminant score function is defined to provide a single indication for classifying input waveforms. The FB is then located from the calculated score maps by sequentially using a threshold, the first local minimum rule of every trace and a median filter. Finally, we adopt synthetic and real shot data examples to demonstrate the effectiveness of CNNs-based waveform classification and FB picking. The results illustrate that CNN is an efficient automatic data-driven classifier and picker.

Yang-Jun Deng;Heng-Chao Li;Lei Pan;Li-Yang Shao;Qian Du;William J. Emery; "Modified Tensor Locality Preserving Projection for Dimensionality Reduction of Hyperspectral Images," vol.15(2), pp.277-281, Feb. 2018. By considering the cubic nature of hyperspectral image (HSI) to address the issue of the curse of dimensionality, we have introduced a tensor locality preserving projection (TLPP) algorithm for HSI dimensionality reduction and classification. The TLPP algorithm reveals the local structure of the original data through constructing an adjacency graph. However, the hyperspectral data are often susceptible to noise, which may lead to inaccurate graph construction. To resolve this issue, we propose a modified TLPP (MTLPP) via building an adjacency graph on a dual feature space rather than the original space. To this end, the region covariance descriptor is exploited to characterize a region of interest around each hyperspectral pixel. The resulting covariances are the symmetric positive definite matrices lying on a Riemannian manifold such that the Log-Euclidean metric is utilized as the similarity measure for the search of the nearest neighbors. Since the defined covariance feature is more robust against noise, the constructed graph can preserve the intrinsic geometric structure of data and enhance the discriminative ability of features in the low-dimensional space. The experimental results on two real HSI data sets validate the effectiveness of our proposed MTLPP method.

Jiawei Zuo;Guangluan Xu;Kun Fu;Xian Sun;Hao Sun; "Aircraft Type Recognition Based on Segmentation With Deep Convolutional Neural Networks," vol.15(2), pp.282-286, Feb. 2018. Aircraft type recognition in remote sensing images is a meaningful task. It remains challenging due to the difficulty of obtaining appropriate representation of aircrafts for recognition. To solve this problem, we propose a novel aircraft type recognition framework based on deep convolutional neural networks. First, an aircraft segmentation network is designed to obtain refined aircraft segmentation results which provide significant details to distinguish different aircrafts. Then, a keypoints’ detection network is proposed to acquire aircrafts’ directions and bounding boxes, which are used to align the segmentation results. A new multirotation refinement method is carefully designed to further improve the keypoints’ precision. At last, we apply a template matching method to identify aircrafts, and the intersection over union is adopted to evaluate the similarity between segmentation results and templates. The proposed framework takes advantage of both shape and scale information of aircrafts for recognition. Experiments show that the proposed method outperforms the state-of-the-art methods and can achieve 95.6% accuracy on the challenging data set.

Yunlong Yu;Fuxian Liu; "Aerial Scene Classification via Multilevel Fusion Based on Deep Convolutional Neural Networks," vol.15(2), pp.287-291, Feb. 2018. One of the challenging problems in understanding high-resolution remote sensing images is aerial scene classification. A well-designed feature extractor and classifier can improve classification accuracy. In this letter, we construct three different convolutional neural networks with different sizes of receptive field, respectively. More importantly, we further propose a multilevel fusion method, which can make judgment by incorporating different levels’ information. The aerial image and two patches extracted from the image are fed to these three different networks, and then, a probability fusion model is established for final classification. The effectiveness of the proposed method is tested on a more challenging data set-AID that has 10000 high-resolution remote sensing images with 30 categories. Experimental results show that our multilevel fusion model gets a significant classification accuracy improvement over all state-of-the-art references.

Jiaojiao Li;Xi Zhao;Yunsong Li;Qian Du;Bobo Xi;Jing Hu; "Classification of Hyperspectral Imagery Using a New Fully Convolutional Neural Network," vol.15(2), pp.292-296, Feb. 2018. With success of convolutional neural networks (CNNs) in computer vision, the CNN has attracted great attention in hyperspectral classification. Many deep learning-based algorithms have been focused on deep feature extraction for classification improvement. In this letter, a novel deep learning framework for hyperspectral classification based on a fully CNN is proposed. Through convolution, deconvolution, and pooling layers, the deep features of hyperspectral data are enhanced. After feature enhancement, the optimized extreme learning machine (ELM) is utilized for classification. The proposed framework outperforms the existing CNN and other traditional classification algorithms by including deconvolution layers and an optimized ELM. Experimental results demonstrate that it can achieve outstanding hyperspectral classification performance.

Peter Planinšič;Dušan Gleich; "Temporal Change Detection in SAR Images Using Log Cumulants and Stacked Autoencoder," vol.15(2), pp.297-301, Feb. 2018. This letter proposes a change detection algorithm for damage assessment caused by fires in Ireland using Sentinel 1 data. The novelty, in this letter, is a feature extraction within tunable <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> discrete wavelet transform (TQWT) using higher order log cumulants of fractional Fourier transform (FrFT), which were fed into a stacked autoencoder (SAE) to distinguish changed and unchanged areas. The extracted features were used to train the SAE layerwise using an unsupervised learning algorithm. After training the decoding layer was replaced by a logistic regression layer to perform supervised fine-tuning and classification. The proposed algorithm was compared with the algorithm that used log cumulants of FrFT within the oriented dual-tree wavelet transform using support vector machine (SVM) classifier. The experimental results showed that the proposed combination of algorithms decreased the overall error (OE) for real synthetic aperture radar images by 6%, when TQWT was used instead of oriented dual-tree wavelet transform and OE was decreased by another 5% when SAE was used instead of the SVM classifier.

Yiming Wang;Xingpeng Mao;Jie Zhang;Yonggang Ji; "Detection of Vessel Targets in Sea Clutter Using In Situ Sea State Measurements With HFSWR," vol.15(2), pp.302-306, Feb. 2018. The detection of vessel targets could be effectively resolved in a high-frequency surface wave radar (HFSWR). However, signals reflected from vessels are concealed by sea clutter in the Doppler spectrum, where such detections are performed. Consequently, differences between these features in the Doppler domain cannot be readily observed, which greatly increases the difficulty in detecting vessel targets. In this letter, in situ sea state information is utilized to facilitate the detection of targets within sea clutter. First, the sea clutter spectrum, which is absent of vessel, is constructed. Second, sensitive sea clutter features that are influenced by vessel targets are selected and analyzed. Third, anomalies in sensitive sea clutter features are detected by obtaining respective thresholds. Finally, vessel targets are identified by the synthesized anomaly detection. Experimental results demonstrate the effectiveness of the proposed method, and the vessels detected using the HFSWR are further verified using synchronous automatic identification system information.

James Theiler;Guen Grosklos; "Corrections to “Problematic Projection to the In-Sample Subspace for a Kernelized Anomaly Detector” [Apr 16 485-489[Name:_blank]]," vol.15(2), pp.307-307, Feb. 2018. In the above paper [1], there are several errors, which we correct here.

* "Introducing IEEE Collabratec," vol.15(2), pp.308-308, Feb. 2018.* Advertisement, IEEE. IEEE Collabratec is a new, integrated online community where IEEE members, researchers, authors, and technology professionals with similar fields of interest can network and collaborate, as well as create and manage content. Featuring a suite of powerful online networking and collaboration tools, IEEE Collabratec allows you to connect according to geographic location, technical interests, or career pursuits. You can also create and share a professional identity that showcases key accomplishments and participate in groups focused around mutual interests, actively learning from and contributing to knowledgeable communities. All in one place! Learn about IEEE Collabratec at ieeecollabratec.org.

* "IEEE Geoscience and Remote Sensing Letters information for authors," vol.15(2), pp.C3-C3, Feb. 2018.* Provides instructions and guidelines to prospective authors who wish to submit manuscripts.

* "IEEE Geoscience and Remote Sensing Letters Institutional Listings," vol.15(2), pp.C4-C4, Feb. 2018.* Advertisements.

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing - new TOC (2018 February 15) [Website]

K. V. Subrahmanyam;Karanam Kishore Kumar;Natalie D. Tourville; "CloudSat Observations of Three-Dimensional Distribution of Cloud Types in Tropical Cyclones," vol.11(2), pp.339-344, Feb. 2018. The present study investigates the three-dimensional distribution of various cloud types in tropical cyclones formed in the North Indian Ocean surrounding the Indian subcontinent using CloudSat observations of 25 cyclones occurred during 2006–2014. A composite cloud type distribution of cirrus (Ci), altostratus (As), altocumulus, stratocumulus, cumulus, nimbostratus and deep convective (DC) is constructed by combining all the observations as a function of the radial distance from the eye of a cyclone for the first time. The present analysis shows that the peak frequency of occurrence of the DC clouds is 50% at ∼50–100 km radial distance from the cyclone's eye. The Ci clouds are found at altitudes around 13–15 km with a maximum frequency of occurrence of 30% at ∼200 km from the center of the cyclone's eye. The present results suggest that there could be a possible discrepancy in classifying the observed clouds into DC and As clouds using CloudSat observations. All the observations during the study period led to the construction of composite cloud type distribution in the tropical cyclones, which aids in visualizing what type of clouds dominates in which part of the cyclone. Thus, the present study provides a three-dimensional distribution of various clouds embedded in tropical cyclones and associated dynamics, which is very important in better representation of tropical cyclones in numerical weather models and can be used to evaluate the tropical cyclone simulations.

Yongming Xu;Anders Knudby;Yan Shen;Yonghong Liu; "Mapping Monthly Air Temperature in the Tibetan Plateau From MODIS Data Based on Machine Learning Methods," vol.11(2), pp.345-354, Feb. 2018. Detailed knowledge of air temperature (<inline-formula><tex-math notation="LaTeX">$T_{a}$</tex-math></inline-formula> ) is desired for various scientific applications. However, in the Tibetan Plateau (TP), the meteorologically observed <inline-formula><tex-math notation="LaTeX">$T_{a}$</tex-math></inline-formula> is limited due to the low density and uneven distribution of stations. This paper aims to develop a 1-km resolution monthly mean <inline-formula> <tex-math notation="LaTeX">$T_{a}$</tex-math></inline-formula> dataset over the TP during 2001–2015 from remote sensing and auxiliary data. 11 environmental variables were extracted from Moderate Resolution Imaging Spectroradiometer (MODIS) data, Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) data and topographic index data. Ten machine learning algorithms were implemented and compared to determine the optimal model for Ta estimation in the TP. The Cubist algorithm outperformed other methods, having the highest accuracy and the lowest sensitivity to cloud contamination. To minimize the overfitting problem, a simple forward variable selection method was introduced and six variables were selected from the original 11 environmental variables. Among these six variables, nighttime land surface temperature (Ts) was the most important predictor, followed by elevation and solar radiance. The seasonal performance of the Cubist model was also assessed. The model had good accuracies in all four seasons, with the highest accuracy in winter (R2 = 0.98 and MAE = 0.63 °C) and the lowest accuracy in summer (R2 = 0.91 and MAE = 0.86 °C). Due to the gaps in MODIS data caused by cloud cover, there were 0.39% missing values in the estimated Ta. To improve the data integrity- Delaunay triangulation interpolation was applied to fill the missing Ta values. The final monthly (2001–2015) Ta dataset had an overall accuracy of RMSE = 1.00 °C and MAE = 0.73 °C. It provides valuable information for climate change assessment and other environmental studies in the TP.

Xiaoxiao Zhang;Zhen-Sen Wu;Xiang Su; "Electromagnetic Scattering from Deterministic Sea Surface With Oceanic Internal Waves via the Variable-Coefficient Gardener Model," vol.11(2), pp.355-366, Feb. 2018. In this paper, a hydrodynamic-electromagnetic model is proposed for analyzing the electromagnetic scattering characteristic of a multiscale deterministic sea surface with internal waves in a two-layer ocean system. A variable-coefficient Gardner model is applied to establish the profile of the internal waves, and the spectrum induced by a variable current is then derived by the action balance equation. The slope-deterministic facet-based two-scale method (SDFb-TSM) is used to calculate the scattering characteristics of the multiscale deterministic sea surface. The total coefficient distribution is then quantitatively obtained from this composite model considering Bragg resonance, tilting effect, and specular reflection effect. Numerical results show that the shape and modulated depth of normalized radar cross section induced by the internal soliton of the Gardner model agree quite well with the measured data. Thus, the Gardner model combined with the SDFb-TSM method performes better than the KdV model combined with statistical scattering methods for the prediction of scattering characteristics of a sea surface with internal waves. The variation in scattering signature characteristics for the depression internal wave are discussed for both one- and two-soliton cases. The influences on the scattering coefficients under different parameters of the internal wave, sea state, and radar system are analyzed. Furthermore, both overtaking-type and head-on-type internal waves are presented and the effect of the perturbed term on internal wave profiles and the scattering characteristics of both one- and two-soliton cases are discussed.

Jin Wang;Jie Zhang;Jing Wang; "Sea Surface Salinity Retrieval Under Rain Based on Aquarius Combined Active/Passive Observations," vol.11(2), pp.367-374, Feb. 2018. Equipped with an L-band radiometer, SMOS, and Aquarius provide an unprecedented sea surface salinity (SSS) dataset of the global oceans in days. The sensitivity of L-band brightness temperature (TB) to SSS variation is about 0.3–0.8 K/psu, which means the salinity signal in TB is very weak. Enormous efforts are devoted to the development, evaluation, and improvement of the SSS retrieval algorithm especially under some unfavorable conditions, i.e., the rain. Rain drops inducing freshening and roughness effects on the sea surface have made the SSS retrieval challenging for years. This paper describes a new method to separate the freshening and roughness effects of rainfall based on the combined active/passive observations of Aquarius. The dependence of the sea surface emissivity (sensitive to both roughness and freshening) on the backscatter (only sensitive to roughness) is obtained and the rain-induced roughness is corrected. The method is applied to the salinity retrieval under rain. The retrieval results ( <inline-formula><tex-math notation="LaTeX">$text{SSS}_{rc}$</tex-math></inline-formula>) are compared with HYCOM data corrected by the rain impact model (<inline-formula><tex-math notation="LaTeX">$text{SSS}_{text{HYCOM}_text{RIM}}$ </tex-math></inline-formula>). The bias of <inline-formula><tex-math notation="LaTeX">$text{SSS}_{text{rc}}$ </tex-math></inline-formula> shows no clear dependence on rain rate. However, the bias of the standard product of Aquarius (<inline-formula><tex-math notation="LaTeX">$text{SSS}_{text{ADPS}}$</tex-math></inline-formula>, V4.0) rises sharply with rain rate. Furthermore, the standard deviation of <inline-formula><tex-math notation="LaTeX"> $text{SSS}_{text{rc}}$</tex-math></inline-formula> is about 0.5 psu, which is also superior to <inline-formula> <tex-math notation="LaTeX">$text{SSS}_{text{ADPS}}$</tex-math></inline-formula> (0.9 psu). Th- above results confirm the feasibility of this new retrieval algorithm for the SSS remote sensing in rainy weather.

Seokhyeon Kim;Kyungrock Paik;Fiona M. Johnson;Ashish Sharma; "Building a Flood-Warning Framework for Ungauged Locations Using Low Resolution, Open-Access Remotely Sensed Surface Soil Moisture, Precipitation, Soil, and Topographic Information," vol.11(2), pp.375-387, Feb. 2018. Soil moisture (SM) plays an important role in determining the antecedent condition of a watershed, while topographic attributes define how and where SM and rainfall interact to create floods. Based on this principle, we present a method to identify flood risk at a location in a watershed by using remotely sensed SM and open-access information on rainfall, soil properties, and topography. The method consists of three hydrologic modules that represent the generation, transfer, and accumulation of direct runoff. To simplify the modeling and provide timely warnings, the flood risk is ascertained based on frequency of exceedance, with warnings issued if above a specified threshold. The simplicity of the method is highlighted by the use of only three parameters for each watershed of interest, with effective regionalization allowing use in ungauged watersheds. For this proof-of-concept study, the proposed model was calibrated and tested for 65 hydrologic reference stations in the Murray–Darling Basin in Australia over a 35-year study period by using satellite-derived surface SM. The three model parameters were first estimated using the first ten-year data and then the model performance was evaluated through flood threshold exceedance analyses over the remaining 25-year study period. The results for estimated parameters and skill scores showed promise. The three model parameters can be regionalized as a function of watershed characteristics, and/or representative values estimated from neighboring watersheds, allowing use in ungauged basins everywhere.

Linwei Yue;Huanfeng Shen;Wei Yu;Liangpei Zhang; "Monitoring of Historical Glacier Recession in Yulong Mountain by the Integration of Multisource Remote Sensing Data," vol.11(2), pp.388-400, Feb. 2018. Yulong Mountain, which is the southernmost snowcapped mountain in mainland Eurasia, has been confronted with significant glacier recession in the last decades due to global climate warming. The recession of these small-scale monsoonal temperate glaciers is a sensitive indicator of global warming. However, there have been few studies that have comprehensively monitored the historical glacier recession in Yulong Mountain area. This paper integrates multisource remote sensing data to monitor the glacial status on Yulong Mountain between 1957 and 2009. Integrating a topographic map, the long-term observed Landsat TM/ETM+ images, and multitemporal digital elevation model datasets, both the area change and regional mass balance of the Yulong glaciers are analyzed. According to the results, the area of the Yulong glaciers decreased from 11.57 to 4.55 km2 at a rate of −0.14 km2 yr−1 over the last 52 years between 1957 and 2009. The 1987–1999 specific mass balance was −0.31 ± 0.33 m yr−1 water equivalent, while the 1987–2008 mass balance was −0.27 ± 0.35 m yr−1 water equivalent. It can be interpreted from the results that the Yulong glaciers have experienced persistent glacier recession during the last decades. The glacier melting is still significant due to the continuously rising temperature. Furthermore, spatially heterogeneous glacier recession has been observed in this area. The glacier changes are spatially varied, which is probably due to the local temperature and precipitation, the glacier sizes, the terminus altitudes, and terrain factors. The influencing elements interacted with each other, and the climate conditions are the dominant factors affecting glacier status.

Ana-Maria Ilisei;Lorenzo Bruzzone; "A Method for Automatic Three-Dimensional Reconstruction of Ice Sheets by Using Radar Sounder and Altimeter Data," vol.11(2), pp.401-415, Feb. 2018. Understanding the processes occurring at the ice sheets requires reliable three-dimensional (3-D) models of the ice sheet geometry. To address this challenge, we propose a technique for the 3-D reconstruction of the ice sheet geometry that uses radar sounder (RS) and altimeter (ALT) data to automatically identify the scale (or grid size) for interpolation. Existing studies derive the interpolation scale empirically, by qualitatively analyzing the RS data sampling and often neglecting the surface topography effects. Our method initially performs the interpolation of RS data at several potential scales. At each scale, it uses the ordinary kriging interpolation method that enables the quantitative analysis of both the RS data sampling and the surface topography. The optimal scale for the estimation of the surface map is identified according to an objective criterion that minimizes the difference to a subset of reference ALT data. Thereafter, the identified optimum scale on the surface is used to estimate the bedrock and ice thickness maps. Thus, the technique is a best-effort approach to the reconstruction of the ice sheet geometry, given the reference surface data and in the absence of reference bedrock data. Results obtained by applying the method to RS and ALT data acquired over the Byrd Glacier in Antarctica, in four regions characterized by different RS sampling and surface topography, confirm its effectiveness. Moreover, they point out that the method could be used for guiding future RS surveys, since the identified optimal scales are typically larger than those needed for addressing specific science objectives.

Dyre Oliver Dammann;Hajo Eicken;Andrew R. Mahoney;Eyal Saiet;Franz J. Meyer;John C. “Craig” George; "Traversing Sea Ice—Linking Surface Roughness and Ice Trafficability Through SAR Polarimetry and Interferometry," vol.11(2), pp.416-433, Feb. 2018. Arctic landfast sea ice is widely utilized for transportation by local communities and industry, with trafficability largely governed by ice roughness. Here, we introduce an approach to evaluate ice roughness that can aid in routing of ice roads and assessment of spatial variability and long-term changes in trafficability. Drawing on synthetic aperture radar (SAR) polarimetry, SAR interferometry (InSAR), and other remote sensing techniques, we integrated approaches into the trafficability assessment that had rarely been applied over sea ice in the past. Analysis of aerial photogrammetry obtained through structure-from-motion helped verify cm-scale accuracy of X-band InSAR-derived ridge height and link L-band polarimetric classification to specific roughness regimes. Jointly, these approaches enable a km-scale evaluation of ridge topography and cm- to m-scale roughness—both critical for the assessment of trafficability. A trafficability index was derived from such SAR data in conjunction with analysis of ice trail routing and ice use near Utqiaġvik, Alaska. The index identifies areas of reduced trafficability, associated with pressure ridges or rubble ice, and served to delineate favorable trail routes for different modes of transportation, with potential uses ranging from ice road routing to emergency evacuation. Community outreach is needed to explore how this approach could assist different ice users in reducing risk, minimizing trail or ice construction efforts, and improving safety.

Jeffrey S. Budge;David G. Long; "A Comprehensive Database for Antarctic Iceberg Tracking Using Scatterometer Data," vol.11(2), pp.434-442, Feb. 2018. This paper describes the development of, and the methodology for, a new, consolidated Brigham Young University (BYU)/National Ice Center (NIC) Antarctic iceberg tracking database. The new database combines daily positional data from the original BYU daily iceberg tracking database derived from scatterometers, and the NIC's weekly Antarctic iceberg tracking database derived mostly from optical and infrared sensors. Interpolation methods and statistical analyses of iceberg locations are discussed. A new, automated method of using positional data and scatterometer backscatter images to estimate sizes and rotational patterns of icebergs is also developed. This information is included in the new database.

Henrique Cândido de Oliveira;Aluir Porfírio Dal Poz;Mauricio Galo;Ayman Fawzy Habib; "Surface Gradient Approach for Occlusion Detection Based on Triangulated Irregular Network for True Orthophoto Generation," vol.11(2), pp.443-457, Feb. 2018. Aerial images of urban areas have been used as base information for a diversity of applications. Considering the great quantity of tall buildings in these areas, it is important to have a method to automatically generate a product called true orthophoto mosaic, which represents all objects above the ground (buildings, bridges, etc.) in their true location. However, to create a true orthophoto, it is necessary to consider the occlusions caused by the surface height variation and to compensate for the lack of information using adjacent aerial images. The automatic occlusion detection is the bottleneck during the true orthophoto mosaic generation. The main aim of this paper is to introduce a new approach for occlusion detection – the surface-gradient-based method (SGBM) applied to a triangulated irregular network (TIN) representation. The originality of the SGBM is the occlusion detection principle, which is based on the concept of surface gradient behavior analysis over a TIN surface. The current methods interpolate a point cloud into a gridded digital surface model, which can introduce artifacts to the representation. The SGBM represents the surface as a TIN-based solid by taking into account the Delaunay constraint in the original point cloud, avoiding the interpolation step. The occlusions are then compensated using specific cost functions and refined via color blending. Experiments were performed and the results were assessed by using quality indicators (completeness), the consistency of orthoimage mosaic, and the time of processing. Experimental results demonstrated the feasibility of the SGBM for occlusion detection in the true orthophoto generation.

Sinong Quan;Boli Xiong;Deliang Xiang;Lingjun Zhao;Siqian Zhang;Gangyao Kuang; "Eigenvalue-Based Urban Area Extraction Using Polarimetric SAR Data," vol.11(2), pp.458-471, Feb. 2018. Urban area extraction using polarimetric synthetic aperture radar (PolSAR) data has become a potential mean to urban studies because it holds the promise that the radar returns from specific scattering characteristics may be emphasized. Due to the high variability of urban land-scape and the existing misdetection of buildings as vegetation, urban area extraction is still a challenging problem. In this paper, an eigenvalue-based urban area extraction method is proposed. First, similar to the entropy/anisotropy plane, a two-dimensional RVI/PA plane is put forward to construct the extractor of buildings with small orientation angles. Second, coupled with the parameters, a robust extractor is introduced to elevate the scattering characteristics of buildings with large orientation angles but to suppress those of others. Finally, data-driven thresholds are investigated and ascertained for the extractors, thus urban areas are extracted. In addition, a change detection-based prescreening method is applied to refine the extraction result. The performance of the proposed method is demonstrated and validated with spaceborne and airborne fully PolSAR data over different test sites. The outputs show that the proposed method provides an overall accuracy of over 90%, as well as better visual results of the extracted buildings.

Zefa Yang;Zhiwei Li;Jianjun Zhu;Axel Preusse;Jun Hu;Guangcai Feng;Yunjia Wang;Markus Papst; "An InSAR-Based Temporal Probability Integral Method and its Application for Predicting Mining-Induced Dynamic Deformations and Assessing Progressive Damage to Surface Buildings," vol.11(2), pp.472-484, Feb. 2018. Surface deformations caused by underground mining are time-dependent and highly nonlinear, and can result in progressive damage to surface structures during underground extraction. However, the previous methods based on the interferometric synthetic aperture radar (InSAR) technique are generally incapable of accurately predicting the mining-induced dynamic deformations occurring during the entire period of underground extraction, due to the inaccurate model parameters inverted under the condition of ignoring the horizontal motions of the InSAR-derived measurements and the model errors for predicting dynamic deformations. Consequently, the risk of mining-related structural damage cannot be reliably assessed based on the deformations predicted by the previous InSAR-based methods. To overcome this limitation, we propose a novel method that combines InSAR with a new mining deformation model: the temporal probability integral method (TPIM). Theoretically, the integration of InSAR and TPIM allows the InSAR-TPIM to accurately predict mining-induced dynamic deformations occurring in the entire period of underground extraction. Furthermore, InSAR-TPIM can reliably assess mining-related progressive structural damage based on the predicted dynamic deformations. However, these advantages cannot be achieved by the previous InSAR-based methods. The Qianyingzi coal mining area of China is selected to test the proposed method. The results demonstrate that the accuracies of the predicted dynamic deformations are about 0.030 and 0.041 m in the horizontal and vertical directions, respectively, which can meet the accuracy requirements of mining-induced dynamic deformation prediction. Furthermore, the comparison between the potential structural damage predicted by the proposed method and the previous InSAR-based methods indicates that the damage risks of around 131 buildings (43.9% of the 298 buildings) are underestimated by the previous InSAR-base- method.

Jan Haas;Yifang Ban; "Urban Land Cover and Ecosystem Service Changes based on Sentinel-2A MSI and Landsat TM Data," vol.11(2), pp.485-497, Feb. 2018. Sustainable development in metropolitan regions is challenging in the light of continuous urbanization. Remote sensing provides timely and reliable information on urban areas and their changing patterns. This study's objectives are to evaluate the contribution of Sentinel-2A (S-2A) data to urban ecosystem service mapping and to investigate spatial ecosystem service characteristics with landscape metrics through a novel method. Service pattern changes between 2005 and 2015 are mapped for Beijing, China. Landscape metrics are used to qualitatively evaluate urban ecosystem service provision bundle changes. S-2A and Landsat TM data are segmented and classified with SVM, distinguishing three artificial and four natural classes based on ecosystem function. Spatial characteristics influencing ecosystem services are quantified with seven landscape metrics. Beijing's urban development is characterized by reduction in agricultural areas in the urban fringe in favor of built-up areas, urban green space, and golf courses. A transformation of old suburban agglomerations into urban green space can be observed. The planar increase in urban areas is accompanied by the creation of managed urban green space. Service bundles based on land cover classes and spatial characteristics decreased more than 30% for bundles that represent food supply, noise reduction, waste treatment, and global climate regulation. Temperature regulation/moderation of climate extremes, recreation/place values/social cohesion, and aesthetic benefits/cognitive development are least affected. This new approach of extending the ecosystem service concept through integration of spatial characteristics of ecosystem service provisional patches through landscape metrics is believed to give a more realistic appraisal of ecosystem services in urban areas.

Kristofer Lasko;Krishna Prasad Vadrevu;Vinh Tuan Tran;Christopher Justice; "Mapping Double and Single Crop Paddy Rice With Sentinel-1A at Varying Spatial Scales and Polarizations in Hanoi, Vietnam," vol.11(2), pp.498-512, Feb. 2018. Paddy Rice is the prevalent land cover in the mosaicked landscape of the Hanoi Capital Region, Vietnam. In this study, we map double and single crop rice in Hanoi using a random forest algorithm and a time-series of Sentinel-1 SAR imagery at 10 and 20 m resolution using VV-only, VH-only, and both polarizations. We compare spatial and areal variation and quantify input band importance, estimate crop growth stages, estimate rice field/collective metrics using Fragstats with image segmentation, and highlight the importance of the results for land use and land cover. Results suggest double crop rice ranged from 208 000 to 220 000 ha with 20-m resolution imagery accounting for the most area in all polarizations. Based on accuracy assessment, we found 10 m data for VV/VH to have highest overall accuracy (93.5%, ±1.33%), while VV at 10 and 20 m had lowest overall accuracies (90.9%, ±1.57; 91.0%, ±2.75). Mean decrease in accuracy suggests for all but VV at 10 m, data from harvest and flooding stages are most critical for classification. Results suggest 20 m data for both VV and VH overestimates rice land cover, however 20 m data may be indicative of rice land use. Analysis of growing season suggests average estimated length of 93–104 days for each season. Commune-level results suggest up to 20% coefficient of variation between VV10m and VH10m with significant spatial variation in rice area. Landscape metrics show rice fields are typically planted in groups of 3–4 fields with over 796 000 collectives and 2.69 million fields estimated in the study area.

Meng Liu;Ronglin Tang;Zhao-Liang Li;Yunjun Yao;Guangjian Yan; "Global Land Surface Evapotranspiration Estimation From Meteorological and Satellite Data Using the Support Vector Machine and Semiempirical Algorithm," vol.11(2), pp.513-521, Feb. 2018. Evapotranspiration (ET) is the combination process of the surface evaporation and plant transpiration, which occur simultaneously, and it links the terrestrial water cycles, carbon cycles, and energy exchange. In this study, based on the observations from 242 global FLUXnet sites, with daily mean temperature, relative humidity, net radiation, wind speed, incoming shortwave radiation, maximum temperature, minimum temperature, normalized difference vegetation index, altitude, difference in temperature, and observed ET as input data, we used a support vector machine and a semiempirical algorithm to estimate the land surface daily ET at nine different vegetation-type sites. Subsequently, based on the meteorological reanalysis data combined with remote sensing data, we estimated regional land surface ET of China during 1982–2010. The results showed that, for all vegetation-type sites, when the predicted ET was validated with the eddy covariance measurements, the support vector machine algorithm undervalued ET while the semiempirical algorithm overvalued ET. When five indicators and the second classification method were selected, the semiempirical algorithm probably could explain 56%–76% of the land surface ET change, whereas the support vector machine algorithm probably could explain 71%–85%. The regional values of annual daily average ET varied from 5.8 to 110.5 W/m2, and the land surface ET overall trend decreased from the southeast to the northwest in China.

Glynn C. Hulley;Nabin K. Malakar;Tanvir Islam;Robert J. Freepartner; "NASA's MODIS and VIIRS Land Surface Temperature and Emissivity Products: A Long-Term and Consistent Earth System Data Record," vol.11(2), pp.522-535, Feb. 2018. Land surface temperature and emissivity (LST&E) determine the total amount of upward long-wave infrared radiation emitted from the Earth's surface, making them key variables in a wide range of studies, including climate variability, land cover/use change, and the energy balance between the land and the atmosphere. LST&E products are currently produced on a routine basis using data from the MODIS instruments on the NASA EOS platforms and by the VIIRS instrument on the Suomi-NPP platform that serves as a bridge between NASA EOS and the next-generation JPSS platforms. Two new NASA LST&E products for MODIS (MxD21) and VIIRS (VNP21) will be produced during 2017 using a new approach that addresses discrepancies in accuracy and consistency between the current suite of MODIS and VIIRS LST split-window-based products. The new approach uses a temperature emissivity separation (TES) algorithm, originally developed for the ASTER instrument, to physically retrieve both LST and spectral emissivity consistently for both sensors with high accuracy and well-defined uncertainties. This study demonstrates continuity between the new MYD21 and VNP21 LST products at the <±0.5 K level, with differences that are invariant to environmental conditions and land cover type. Furthermore, MYD21 and VNP21 retrieved emissivities matched closely in magnitude and temporal variation to within 1%–2% over two land validation sites consisting of quartz sands and grassland. Continuity between the new suite of MODIS and VIIRS LST&E products will ensure a consistent and well-characterized long-term LST&E data record for better monitoring and understanding trends in Earth system behavior.

Wei Ao;Feng Xu;Yongchen Li;Haipeng Wang; "Detection and Discrimination of Ship Targets in Complex Background From Spaceborne ALOS-2 SAR Images," vol.11(2), pp.536-550, Feb. 2018. This paper proposes a novel method for ship detection and discrimination in complex background from synthetic aperture radar (SAR) images. It first implements a pixel-level land–sea segmentation with the aid of a global 250-m water mask. Then, an efficient multiscale constant false alarm rate (CFAR) detector with generalized Gamma distribution clutter model is designed to detect candidate targets in the sea. At last, eigenellipse discrimination and maximum-likelihood (ML) discrimination are designed to further exclude false alarm nonship objects in nearshore and harbor area. The proposed land–sea segmentation method is compared with multilevel Otsu method. The proposed multiscale ship detector is compared with conventional CFAR detectors. These contrast experiments show the good performance of our method. Finally, experiments undertaken on actual ALOS-2 SAR data show the efficacy of the proposed approach in detecting nearshore ship targets in a complex coastal environment.

Zu-Zhen Huang;Ze-Gang Ding;Jia Xu;Tao Zeng;Li Liu;Zhi-Rui Wang;Chang-Hui Feng; "Azimuth Location Deambiguity for SAR Ground Moving Targets via Coprime Adjacent Arrays," vol.11(2), pp.551-561, Feb. 2018. The ground moving target's radial velocity estimation based on the interferometric phase in a multichannel synthetic aperture radar (SAR) system suffers from 2π modulo folding, resulting in the target's azimuth location ambiguity in the SAR image. To address this problem, a novel coprime adjacent arrays SAR (CAA-SAR) is proposed in this paper. The two sparse uniform subarrays constituting the CAA are arranged adjacently with a conjunct element, then a virtual array can be obtained with much more virtual elements and smaller element spacing, which will help to solve the azimuth location ambiguity. After ground clutter suppression, multiple pixels’ samplings are utilized based on the MUSIC algorithm for estimating radial velocity as well as azimuth shift by exploiting all virtual elements of CAA-SAR. Compared with the existing nonuniform linear array SAR method based on Chinese remainder theorem, the CAA-SAR can obtain a better accuracy with the same number of elements in a larger sparse configuration, or use fewer elements to obtain an approximate accuracy in almost the same physical aperture. Compared with the coprime arrays SAR, the CAA-SAR has a better accuracy due to its larger number of unique virtual elements and longer physical aperture. Finally, some results of numerical experiments are provided to demonstrate the effectiveness of the proposed method.

Sithara Kanakaraj;Madhu S. Nair;Saidalavi Kalady; "SAR Image Super Resolution using Importance Sampling Unscented Kalman Filter," vol.11(2), pp.562-571, Feb. 2018. Synthetic aperture radar (SAR) imaging is a crucial tool in providing images of the earth's surface for military and civilian applications such as target surveillance and its classification. The precision of the application degrades with the presence of inherent speckle and poor resolution of images from the SAR image acquisition devices. Thus, the objective of the proposed method is to develop a technique to enhance the resolution while despeckling the inherent noise, simultaneously, since the conventional super resolution methods have failed to do the same. Moreover, the works from the literature that super resolve SAR images have also neglected the signal-dependent noise model. The work proposed in this paper significantly reduces the speckle and super resolves the SAR image using an Importance Sampling Unscented Kalman Filter framework that best models the non-linearity of the system. The technique has been assessed quantitatively and qualitatively on synthetic images as well as on real SAR images. The performance evaluation based on peak signal-to-noise-ratio, structural similarity index measure, feature similarity index measure, edge preservation factor, and equivalent number of looks values throw light on the superiority of the proposed method over the standard and other recent techniques. This can serve to generate images with a better reconstructive quality that would aid various applications in multidisciplinary domains.

Ran Liu;Wenkai Li;Xiaoping Liu;Xingcheng Lu;Tianhong Li;Qinghua Guo; "An Ensemble of Classifiers Based on Positive and Unlabeled Data in One-Class Remote Sensing Classification," vol.11(2), pp.572-584, Feb. 2018. One-class remote sensing classification refers to the situations when users are only interested in one specific land type without considering other types. The positive and unlabeled learning (PUL) algorithm, which trains a binary classifier from positive and unlabeled data, has been shown to be promising in one-class classification. The implementation of PUL by a single classifier has been investigated. However, implementing PUL using multiple classifiers and creating classifier ensembles based on PUL have not been studied. In this research, we investigate the implementations of PUL using several classifiers, including generalized linear model, generalized additive model, multivariate adaptive regression splines, maximum entropy, backpropagation neural network, and support vector machine, as well as three ensemble methods based on majority vote, weighted average, and weighted vote combination rules. These methods are applied in classifying the urban areas from four remote sensing imagery of different spatial resolutions, including aerial photograph, Landsat 8, WorldView-3, and Gaofen-1. Experimental results show that classifiers can successfully extract the urban areas with high accuracies, and the ensemble methods based on weighted average and weighted vote generally outperform the individual classifiers on different datasets. We conclude that PUL is a promising method in one-class remote sensing classification, and the classifier ensemble based on PUL can significantly improve the accuracy.

Nan Mo;Ruixi Zhu;Li Yan;Zhan Zhao; "Deshadowing of Urban Airborne Imagery Based on Object-Oriented Automatic Shadow Detection and Regional Matching Compensation," vol.11(2), pp.585-605, Feb. 2018. With the increase of the spatial resolution of aerial images, the shadow problem is more prominent. The shadows affect the applications such as object recognition, image dense matching, and object classification. Existing shadow detection methods can acquire results with high accuracy, but usually need much manual intervention and the traditional stretch models generally lead to color distortion and undercompensation. Therefore, we propose an object-oriented automatic shadow detection method without manual intervention and a shadow compensation method by regional matching. In the proposed method, pixel-based soft shadow detection, which uses Gaussian mixture model to simulate the gray distribution and refines soft shadow map with guiled filtering, is combined with image segmentation result to obtain accurate shadow regions with complete shape and no hole. Then shadow regions are compensated, with less loss of details and brightness imbalance, referring to their optimal homogeneous nonshadow region obtained by regional matching based on Bag-of-Words. The total variation model is used to decrease the noise amplified by the pixel-based stretch and boundary effect in the result. Experiments are performed on three publicly available high-resolution aerial images to demonstrate the superiority of our proposed methods. It shows that the proposed method can accurately detect shadows from urban high-resolution aerial images with an overall rate of over 88%. The compensation results display excellent visual effects compared with the state-of-the-art methods, consistent with the true color of ground objects.

Ruisheng Wang;Jiju Peethambaran;Dong Chen; "LiDAR Point Clouds to 3-D Urban Models$:$ A Review," vol.11(2), pp.606-627, Feb. 2018. Three-dimensional (3-D) urban models are an integral part of numerous applications, such as urban planning and performance simulation, mapping and visualization, emergency response training and entertainment, among others. We consolidate various algorithms proposed for reconstructing 3-D models of urban objects from point clouds. Urban models addressed in this review include buildings, vegetation, utilities such as roads or power lines and free-form architectures such as curved buildings or statues, all of which are ubiquitous in a typical urban scenario. While urban modeling, building reconstruction, in particular, clearly demand specific traits in the models, such as regularity, symmetry, and repetition; most of the traditional and state-of-the-art 3-D reconstruction algorithms are designed to address very generic objects of arbitrary shapes and topology. The recent efforts in the urban reconstruction arena, however, strive to accommodate the various pressing needs of urban modeling. Strategically, urban modeling research nowadays focuses on the usage of specialized priors, such as global regularity, Manhattan-geometry or symmetry to aid the reconstruction, or efficient adaptation of existing reconstruction techniques to the urban modeling pipeline. Aimed at an in-depth exploration of further possibilities, we review the existing urban reconstruction algorithms, prevalent in computer graphics, computer vision and photogrammetry disciplines, evaluate their performance in the architectural modeling context, and discuss the adaptability of generic mesh reconstruction techniques to the urban modeling pipeline. In the end, we suggest a few directions of research that may be adopted to close in the technology gaps.

Xiaolong Cheng;Xiaojun Cheng;Quan Li;Liwei Ma; "Automatic Registration of Terrestrial and Airborne Point Clouds Using Building Outline Features," vol.11(2), pp.628-638, Feb. 2018. Terrestrial laser scanner (TLS) and airborne laser scanner (ALS) can effectively capture point clouds from side or top view, respectively. Registering point clouds captured by ALS and TLS provides an integrated data source for three-dimensional (3-D) reconstruction. However, registration is difficult between TLS and ALS data because of the differences in scanning perspectives, scanning area, and spatial resolutions. A new method that can achieve automatic horizontal registration with ALS and TLS data based on building contour features is proposed in this study. The key steps include horizontal and vertical registrations based on 2-D building outlines and ground planes in ALS and TLS data, respectively. First, the 2-D building outlines are extracted from both ALS and TLS data. Second, the horizontal registration is accomplished by using the four-point congruent sets method for initial registration and the global optimization method for refined registration. Finally, the ground surface in the same region of ALS and TLS data are fitted for vertical registration, and the average elevation difference between the corresponding ground planes is calculated as the translation parameter value in the vertical direction. The results indicate that the proposed method can successfully match ALS and TLS data with an accuracy of 0.2-m both in the horizontal and vertical directions.

Benjamin J. Southwell;Andrew G. Dempster; "A New Approach to Determine the Specular Point of Forward Reflected GNSS Signals," vol.11(2), pp.639-646, Feb. 2018. Accurate determination of the specular point is important for simulating and processing reflected global navigation satellite system signals for remote sensing applications. Existing methods for determining the specular point are based on both spherical and ellipsoidal approximations of the Earth and employ either Snell's law or Fermat's principle to formulate the problem. By analysis and simulation, it is shown that these methods produce significant errors at intermediate latitudes. In this paper, a novel formulation for the solution of the specular point is proposed that satisfies Snell's Law on the WGS84 ellipsoid. The proposed method is compared with the existing methods for various receiver orbit configurations and algorithm augmentations. It is shown that the proposed method is more accurate than the existing methods and more computationally efficient than the minimum path length (MPL) method. Additionally, the resultant grazing angles, MPL method errors, and specular point locations as a function of the receiver orbit are investigated, leading to the finding that the likelihood of an error is significant for geometries favored by reflectometry applications.

Xinglin Lu;Ao Song;Rongyi Qian;Lanbo Liu; "Anisotropic Reverse-Time Migration of Ground-Penetrating Radar Data Collected on the Sand Dunes in the Badain Jaran Desert," vol.11(2), pp.647-654, Feb. 2018. Ground-penetrating radar (GPR) profiling is the primary tool to provide detail information of the internal structure and characteristics inside a sand dune in the desert area. However, with the severe elevation change in a short horizontal distance and on the rugged and complex surface topography of the sand dunes, getting clear imaging of the internal structure of sand dunes from the GPR profiling is still a big challenge. The classic imaging technique such as the Kirchhoff migration has been applied to process GPR data on sand dunes, with limited success. The reverse-time migration (RTM) technique is the most advanced imaging technique that can handles GPR data acquired on rugged surface with severe topographic relief to generate subsurface structural images with high fidelity and has been tried to process GPR data on sand dunes. The results are encouraging, and the imaging quality is significantly improved. The finite-difference time-domain (FDTD) method is the major numerical tool for forward and backward continuations of the wave field during the RTM process. There are two aspects for using FDTD in RTM of GPR data on sand dunes still need improvement: the numerical scattering caused by the staircase approximation of the ground surface by using gridding in the Cartesian coordinate, and the negligence of the possible anisotropy of the electromagnetic material properties due to the calcareous cementation bedding inside a sand dune. In this paper, we develop the RTM algorithm based on the staggered grid FDTD that handles the rugged topographic surface by using the curvilinear coordinate, and the possible anisotropic radar wave velocity of the sand dune media. We first demonstrate the equivalency of the nonuniform, isotropic medium and the uniform, anisotropic medium for justifying using the uniform, anisotropic velocity in the RTM by synthetic modeling. Next, we validate our approach of using the synthetic data with the comparison of using the Cartesian coordinate- and the curvilinear coordinate in an isotropic medium. The results indicate that the RTM algorithm using the curvilinear coordinate can efficiently eliminate the adverse effect of the staircase approximated boundary of the topography surface. Finally, we processed the real GPR data collected on a sand dune in the Badain Jaran desert by using the curvilinear coordinate and the uniform, anisotropic velocity in FDTD forward and backward wave field continuation. Comparison of the results indicates that the RTM imaging using the boundary-conforming curvilinear coordinate and anisotropic velocity gains more coherent and higher resolution images for the calcareous cementation layers and the water table.

Ling Zhang;Zhaofa Zeng;Jing Li;Jingyi Lin;Yingsa Hu;Xuegang Wang;Xiaodong Sun; "Simulation of the Lunar Regolith and Lunar-Penetrating Radar Data Processing," vol.11(2), pp.655-663, Feb. 2018. Lunar-penetrating radar (LPR) was conducted by the “Yutu” rover of China's Chang-E 3 lunar mission to study the shallow subsurface of the Moon. Both regolith modeling and numerical simulation can provide a reliable reference for data processing of the Moon. In this study, a 3-D lunar regolith model that considered some key factors including terrain, rocks, randomness of medium, and permittivity change with depth is built. Based on LPR numerical simulation, v(z) f-k migration, with high accuracy for vertical velocity variations, is carried out. Compared with Stolt f-k migration, which is limited to constant velocity, v(z) f-k migration performs better. We have designed a workflow for LPR data of Chang-E 3 mission, such as v(z) f-k migration, filters, and gain. Finally, according to the LPR real data result, we estimate the thickness of the regolith, the location and physical parameters of several rocks, and randomness of medium. Besides, the present study provides a good reference for further understanding of lunar near-surface geological information.

Albéric De Coster;Sébastien Lambot; "Fusion of Multifrequency GPR Data Freed From Antenna Effects," vol.11(2), pp.664-674, Feb. 2018. Several data fusion approaches have been developed to optimize both resolution and characterization depth for multifrequency ground-penetrating radar (GPR). In this study, we propose a novel physically based method to merge radar data coming from antennas operating in different frequency ranges. The strategy relies on the removal of the source and antenna effects from GPR data and the subsequent fusion of the resulting signals, which are now normalized, in the frequency domain. The approach used to filter out antenna effects resorts to an intrinsic, closed-form solution of Maxwell's equations to describe the radar-antenna-medium system. We validated the multifrequency GPR data fusion approach through laboratory experiments with measurements performed in far- and near-field conditions above a copper plane and pipes buried at different depths in a sandbox. The results demonstrated the benefit of filtering the frequency-dependent antennas effects before data fusion. Enhanced radargrams were subsequently obtained as a result of the broadening of the spectral bandwidth. This physically based fusion approach appears to be very promising to improve subsurface imaging.

IEEE Geoscience and Remote Sensing Magazine - new TOC (2018 February 15) [Website]

* "[Front cover[Name:_blank]]," vol.5(4), pp.C1-C1, Dec. 2017.* Presents the front cover for this issue of the publication.

* "IEEE Geoscience and Remote Sensing Magazine," vol.5(4), pp.C2-C2, Dec. 2017.* Prospective authors are requested to submit new, unpublished manuscripts for inclusion in the upcoming event described in this call for papers.

* "Table of contents," vol.5(4), pp.1-2, Dec. 2017.* Presents the table of contents for this issue of the periodical.

* "Staff List," vol.5(4), pp.2-2, Dec. 2017.* Provides a listing of current staff, committee members and society officers.

Lorenzo Bruzzone; "A Farewell from the Founding Editor-in-Chief [From the Editor[Name:_blank]]," vol.5(4), pp.3-4, Dec. 2017. Presents the farewell message from the founding Editor-in-Chief.

* "BGDDS 2018," vol.5(4), pp.4-4, Dec. 2017.* Describes the above-named upcoming conference event. May include topics to be covered or calls for papers.

* "CISS 2018," vol.5(4), pp.4-4, Dec. 2017.* Describes the above-named upcoming conference event. May include topics to be covered or calls for papers.

Adriano Camps; "Working Toward a Brighter Future for the GRSS [President's Message[Name:_blank]]," vol.5(4), pp.5-7, Dec. 2017. Presents the President’s message for this issue of the publication.

Xiao Xiang Zhu;Devis Tuia;Lichao Mou;Gui-Song Xia;Liangpei Zhang;Feng Xu;Friedrich Fraundorfer; "Deep Learning in Remote Sensing: A Comprehensive Review and List of Resources," vol.5(4), pp.8-36, Dec. 2017. Central to the looming paradigm shift toward data-intensive science, machine-learning techniques are becoming increasingly important. In particular, deep learning has proven to be both a major breakthrough and an extremely powerful tool in many fields. Shall we embrace deep learning as the key to everything? Or should we resist a black-box solution? These are controversial issues within the remote-sensing community. In this article, we analyze the challenges of using deep learning for remote-sensing data analysis, review recent advances, and provide resources we hope will make deep learning in remote sensing seem ridiculously simple. More importantly, we encourage remote-sensing scientists to bring their expertise into deep learning and use it as an implicit general model to tackle unprecedented, large-scale, influential challenges, such as climate change and urbanization.

Pedram Ghamisi;Naoto Yokoya;Jun Li;Wenzhi Liao;Sicong Liu;Javier Plaza;Behnood Rasti;Antonio Plaza; "Advances in Hyperspectral Image and Signal Processing: A Comprehensive Overview of the State of the Art," vol.5(4), pp.37-78, Dec. 2017. Recent advances in airborne and spaceborne hyperspectral imaging technology have provided end users with rich spectral, spatial, and temporal information. They have made a plethora of applications feasible for the analysis of large areas of the Earth?s surface. However, a significant number of factors-such as the high dimensions and size of the hyperspectral data, the lack of training samples, mixed pixels, light-scattering mechanisms in the acquisition process, and different atmospheric and geometric distortions-make such data inherently nonlinear and complex, which poses major challenges for existing methodologies to effectively process and analyze the data sets. Hence, rigorous and innovative methodologies are required for hyperspectral image (HSI) and signal processing and have become a center of attention for researchers worldwide.

Jordi Munoz-Mari;Emma Izquierdo-Verdiguier;Manuel Campos-Taberner;Adrian Perez-Suay;Luis Gomez-Chova;Gonzalo Mateo-Garcia;Ana B. Ruescas;Valero Laparra;Jose A. Padron;Julia Amoros-Lopez;Gustau Camps-Valls; "HyperLabelMe : A Web Platform for Benchmarking Remote-Sensing Image Classifiers," vol.5(4), pp.79-85, Dec. 2017. HyperLabelMe is a web platform that allows the automatic benchmarking of remote-sensing image classifiers. To demonstrate this platform's attributes, we collected and harmonized a large data set of labeled multispectral and hyperspectral images with different numbers of classes, dimensionality, noise sources, and levels. The registered user can download training data pairs (spectra and land cover/use labels) and submit the predictions for unseen testing spectra. The system then evaluates the accuracy and robustness of the classifier, and it reports different scores as well as a ranked list of the best methods and users. The system is modular, scalable, and ever-growing in data sets and classifier results.

Werner Wiesbeck;Martti Hallikainen;Jasmeet Judge; "IGARSS in Fort Worth: Impressions from the First Days [Conference Reports[Name:_blank]]," vol.5(4), pp.86-93, Dec. 2017. Presents information on the IGARSS 2017 conference.

Martti Hallikainen;Werner Wiesbeck;Paolo Gamba;Jasmeet Judge;Xiuping Jia; "Awards Presented at the IGARSS 2017 Banquet [Conference Reports[Name:_blank]]," vol.5(4), pp.94-107, Dec. 2017. Presents the recipients of GRSS society awards presented at the IGARSS 2017 Conference.

Feng Xu;Ya-Qiu Jin; "Remote Sensing with Intelligent Processing 2017 in Shanghai, China [Conference Reports[Name:_blank]]," vol.5(4), pp.108-123, Dec. 2017. Presents information on the Remote Sensing with Intelligent Processing 2017 conference.

Devis Tuia;Gabriele Moser;Bertrand Le Saux;Benjamin Bechtel;Linda See; "The 2017 IEEE Geoscience and Remote Sensing Society Data Fusion Contest: Open Data for Global Multimodal Land Use Classification [Technical Committees[Name:_blank]]," vol.5(4), pp.110-114, Dec. 2017. Presents information on the 2017 IEEE Geoscience and Remote Sensing Society Data Fusion Contest.

Emmett Ientilucci; "GRSS Western New York Chapter Status and Activities 2016-2017 [Chapters[Name:_blank]]," vol.5(4), pp.115-118, Dec. 2017. Presents information on various GRS Society chapters.

Nariman Firoozy;Dario Schor; "GRSS Winnipeg Chapter Overview [Chapters[Name:_blank]]," vol.5(4), pp.119-120, Dec. 2017. Presents information on various GRS Society chapters.

* "GRSS Members Elevated to IEEE Senior Member in August 2017 [GRSS Member Highlights[Name:_blank]]," vol.5(4), pp.121-121, Dec. 2017.* Presents GRSS members who were elevated to the status of IEEE Senior Member.

* "[Calendar[Name:_blank]]," vol.5(4), pp.122-122, Dec. 2017.* Provides a notice of upcoming conference events of interest to practitioners and researchers.

* "2017 Index IEEE Geoscience and Remote Sensing Magazine Volume 5," vol.5(4), pp.124-128, Dec. 2017.* This index covers all technical items - papers, correspondence, reviews, etc. - that appeared in this periodical during the year, and items from previous years that were commented upon or corrected in this year. Departments and other items may also be covered if they have been judged to have archival value. The Author Index contains the primary entry for each item, listed under the first author's name. The primary entry includes the co-authors' names, the title of the paper or other item, and its location, specified by the publication abbreviation, year, month, and inclusive pagination. The Subject Index contains entries describing the item under all appropriate subject headings, plus the first author's name, the publication abbreviation, month, and year, and inclusive pages. Note that the item title is found only under the primary entry in the Author Index.

Topic revision: r6 - 22 May 2015, AndreaVaccari
 
banner.png
©2018 University of Virginia. Privacy Statement
Virginia Image and Video Analysis - School of Engineering and Applied Science - University of Virginia
P.O. Box 400743 - Charlottesville, VA - 22904 - E-Mail viva.uva@gmailREMOVEME.com