The registration of optical and synthetic aperture radar (SAR) images is severely affected by nonlinear radiometric distortions (NRD) and speckle noise. To address these challenges, we propose a novel Multi-scale Phase Asymmetry based Optical SAR registration (MSPA-OS) method, which pioneeringly incorporates phase asymmetry (PA) into the feature extraction process. Compared with phase congruency (PC), PA is more robust to noise. By aggregating PA across multiple scales, we efficiently extract the comprehensive structural features of images. Moreover, a multi-region cross-scale matching (MRCSM) strategy with the rotation-invariant descriptors is devised to handle substantial geometric deformations. Furthermore, MSPA-OS employs a set of monogenic filters to process images, significantly increasing the computational speed. Finally, we compare the performance of MSPA-OS with those of seven state-of-the-art methods using synthetic and real datasets. The experimental results show that MSPA-OS exhibits competitive registration robustness and speed.

Detecting tiny infrared (IR) targets in diverse complex backgrounds faces many challenges, e.g., extremely few features of the tiny targets, cluttered backgrounds, and interferences from surrounding similar objects. In this letter, we propose a novel size-aware local contrast measure (SALCM) method to detect tiny IR targets. First, to tackle the problem of extremely few features, various local features are extracted through monogenic signal decomposition, which can effectively enrich the potential features of the tiny targets. Second, the Canny detector is used to precisely delineate the contours of multiple candidate targets in the fused image to estimate the exact shapes and sizes of candidate targets. This ensures that the proposed method adapts to both tiny targets and small targets (with relatively larger sizes). Finally, local contrast enhancement is used to highlight the target regions while suppressing the background clutters and interferences from surrounding similar objects, leading to accurate detection. The experimental results on six real IR target datasets demonstrate the superiority of the proposed method in terms of target enhancement, background suppression, and detection accuracy, for detecting IR targets of various sizes.

Detection and tracking using video synthetic aperture radar (ViSAR) have attracted a great deal of attention in recent years due to its ability to produce high-resolution videos for regions of interest. In this work, we have chosen the background-aware correlation filter (BACF) as a key algorithm due to its superior performance in real-time tracking scenarios. Dim and small shadows with weak visual features, time-varying shadows, and complex environments pose serious challenges to the target tracking problem using ViSAR videos. These factors lead to a multipeak problem in the BACF and many other correlation filter-based algorithms. As a result, incorrect target detection, template model contamination, and tracker drift or failure during long-term tracking can occur. In this work, we propose a novel appearance-distance information-assisted probabilistic data association (AD-PDA) algorithm to tackle the multipeak problem. Based on the correlation outputs of the BACF, the AD-PDA algorithm selects multiple peak locations as validated measurements. By exploiting the appearance-distance probability distribution functions, the AD-PDA algorithm calculates the mixed appearance-distance association weights and estimates target states accurately. Furthermore, we propose an efficient AD-PDA-BACF algorithm that can track targets accurately by combining the AD-PDA and BACF algorithms. This study conducts experiments using two public ViSAR video datasets released by the Sandia National Laboratory. Our results demonstrate that the proposed algorithm outperforms several state-of-the-art algorithms in tracking dim and small targets in terms of tracking accuracy, success rate, and tracking speed.

Deep neural networks have been successfully applied to synthetic aperture radar automatic target recognition (SAR-ATR). The high performance of related methods relies on a large amount of training data. However, in practical applications, collecting SAR data is an expensive and time-consuming process, which results in few data available for training. Recognition with limited training data has become an essential issue in SAR-ATR. To solve this problem, we propose a novel graph metalearning method, in which the graph neural network is combined with metalearning to recognize SAR targets using few labeled SAR images. First, we use simulated SAR data to acquire metaknowledge in the form of feature extractor parameters. The feature extractor is then employed to encode the labeled and unlabeled real SAR images into embedding vectors. Second, the embedding vectors are constructed as a fully connected graph, where each node represents an image and each edge is the similarity between two nodes. After iteratively updating the graph by neighborhood aggregation, the new representations of the nodes and their relationship are obtained. Finally, we generate a prediction distribution on the target classes by fusing the node and edge information of the unlabeled image. The experimental results demonstrate the proposed method’s superiority in recognition accuracy with few training data.

Video synthetic-aperture radar (ViSAR) can obtain high-resolution images of a region of interest at a high frame rate. This feature of ViSAR is helpful for real-time detection and tracking of moving targets. Moving-target tracking using ViSAR images is a typical dim-target-tracking problem. In the context of this article, dim targets correspond to the shadows of the moving vehicles cast onto the stationary background scene, which appear at lower gray levels compared with the background clutter. To detect and track multiple slowly maneuvering targets in the ViSAR imagery, we propose a novel algorithm, the expanding and shrinking strategy-based particle filter/dynamic programming-based track-before-detect (ES-TBD) algorithm. To the best of our knowledge, our work represents the first algorithm to deal with the ViSAR-detection and tracking problem using the TBD method. Furthermore, to detect and track a time-varying number of targets, we also propose a novel region-partitioning-based ES-TBD (RP-TBD) algorithm. By exploiting the common information shared between the batches of measurement data and the modeling merit-function-integrated particle filters (PFs), the RP-TBD partitions the observation region into a predicted subregion and an innovative subregion. The RP-TBD algorithm detects newborn targets in the innovative subregion, while maintains tracks of known targets in the predicted subregion. Experimental results using real ViSAR images show that the proposed algorithms outperform the state-of-the-art algorithms on detecting and tracking multiple dim targets in terms of location accuracy and false-alarm suppression.

Data integrity and correctness are essential for many internet of things applications. This work addresses the data completion problem for long-term operating, large-scale and heterogeneous wireless sensor networks (WSNs), using tensor completion technique. In most previous works, the data models only involve single temporal and single spatial dimensions, ignoring the higher-order intrinsic structures. Besides, most existing schemes use low-rankness and sparsity in a fixed domain to describe the data correlation, but neither can fit well with all the temporal, spatial and attribute correlation. To address these issues, we first propose a higher-order tensor model for data from WSNs, namely multilinear plus sparse (MPS) model, which has multiple temporal, multiple spatial and one attribute modes. The core idea is to describe the temporal-spatial-attribute correlation by the sparsity in the multilinear transform domain. Appropriate sparsifying transforms can be flexibly determined according to the specific natures of each mode. For the temporal and spatial modes, analytical transforms are adopted considering computational efficiency and addressing continuity feature. For the attribute mode, we utilize MPS to propose a robust ℓ0-norm based principle component analysis (robust ℓ0-PCA) algorithm to adaptively learn the sparsifying transform for various heterogeneous networks. Finally, based on MPS and the learned transform, we develop an efficient tensor completion algorithm using alternating direction method of multipliers, namely MPS-TC. The experimental results on real-life WSN data verify that the proposed robust ℓ0-PCA algorithm can learn an attribute transform more robustly than the conventional PCA and ℓ1-PCA algorithms; and the proposed MPS-TC algorithm outperforms the state-of-the-art tensor completion algorithms in terms of signal-to-noise ratio and root mean square error.

We address the sparse signal reconstruction in a wireless sensor network (WSN) via distributed compressed sensing (DCS). The multiple sparse signals from WSN are modeled by the mixed-support set (MSM) model, which describes the inter-correlation of the signals by the common support set and represents the individual features by the innovation support sets. We propose a novel common subspace pursuit (CSP) algorithm to estimate the common support set, in order to reduce the reconstruction error and computing time. The results of simulations on a hierarchical clustering based WSN show that the proposed CSP algorithm is superior over the conventional algorithms in terms of both reconstruction error and runtime.

With the increase of the imaging resolution, the resulting enormous amount of sampling raw data aggravates transmission and storage load for multi-channel synthetic aperture radar (SAR) system. Considering the fact that the correlation among the dual-channel SAR images is high, we propose a Bayesian compressive sensing (BCS) based SAR imaging algorithm for ground moving targets indication (GMTI) system, which uses Laplace priors on the basis coefficients in a hierarchical manner. The simulation results show that the proposed algorithm can successfully detect the moving target and meanwhile suppress the static clutter scattering centers, with 50% sampling data of those required by the range-Doppler (RD) algorithm.

Bearing-only filtering algorithms used in submarine tracking commonly assume that the target and ownship move in the same plane. In real-world scenarios, the target and ownship may actually move in different planes, since it may be advantageous for the ownship to do so. Tracking of the target for this scenario can be accomplished by using passive bearing and elevation measurements. Then the algorithm for angle-only filtering in 3D used in passive ranging with an infrared search and track sensor can be applicable to the submarine tracking problem. Advances in sensor technology and signal processing would allow a future sensor system to collect bearing and elevation measurements by an ownship with improved accuracy. In this paper, we analyze the tracking accuracy of a submarine tracking scenario by using angle-only filtering in 3D while varying the height difference between the target and ownship planes. We use an extended Kalman filter, unscented Kalman filter, range-parametrized UKF, and particle filter to compare the state estimation accuracy. Our results show that the height of the target can be estimated accurately by the proposed algorithms for small height differences between the target and ownship.

This paper presents a compressed sensing based sensor selection algorithm for direction-of-arrival estimation, in a large scale randomly distributed sensor array. First, a target tracker is employed for the prior information of target position. Second, according to the prior information, we reduce rank of sensing matrix by narrowing the interesting area. Third, a linear independence combination of sensors is utilized to represent the whole sensor array through eigendecomposition. Finally, the statistic and dynamic simulations are carried out, which demonstrate the feasibility of the proposed algorithm.