Multi-aspect Synthetic Aperture Radar Research Articles

Multi-Aspect SAR Target Recognition Based on Non-Local and Contrastive Learning

Synthetic aperture radar (SAR) automatic target recognition (ATR) has been widely applied in multiple fields. However, the special imaging mechanism of SAR results in different visual features of the same target at different azimuth angles, so single-aspect SAR target recognition has the limitation of observing the target from a single perspective. Multi-aspect SAR target recognition technology can overcome this limitation by utilizing information from different azimuths and effectively improve target recognition performance. Considering the order dependency and data limitation of existing methods, this paper proposes a multi-aspect SAR recognition method based on Non-Local, which applies a self-attention calculation to feature maps to learn the correlation between multi-aspect SAR images. Meanwhile, in order to improve the generalization ability of the proposed method under limited data, a network based on contrastive learning was designed to pre-train the feature extraction part of the whole network. The experimental results using the MSTAR dataset show that the proposed method has excellent recognition accuracy and good robustness.

3-D Stereo Geolocation of Radar Reflectors Using Multiaspect SAR Acquisitions

With well-controlled orbits and a more accurate timing system of the new-generation synthetic aperture radar (SAR) satellites, the ability of high-precision geolocation with multiple SAR acquisitions has been verified based on radar reflectors. In this letter, SAR absolute and differential geolocation methods are introduced and implemented to retrieve the coordinates of two different types of radar reflectors: corner reflectors (CRs) and multidirectional dihedral reflectors (MDRs). The experimental results demonstrate that decimeter- even centimeter-level positioning accuracy can be obtained by SAR stereo geolocation based on multiaspect SAR acquisitions from RadarSAT-2 (RS-2), COSMO-SkyMed (CSK), Sentinel-1 (S1), and TerraSAR-X (TSX), respectively.

A Probabilistic Approach for Stereo 3D Point Cloud Reconstruction from Airborne Single-Channel Multi-Aspect SAR Image Sequences

We investigate the problem of obtaining dense 3D reconstruction from airborne multi-aspect synthetic aperture radar (SAR) image sequences. Dense 3D reconstructions of multi-view SAR images are vulnerable to anisotropic scatters. To address this issue, we propose a probabilistic 3D reconstruction method based on jointly estimating the pixel’s height and degree of anisotropy. Specifically, we propose a mixture distribution model for the stereo-matching results, where the degree of anisotropy is modeled as an underlying error source. Then, a Bayesian filtering method is proposed for dense 3D point cloud generation. For real-time applications, redundancy in multi-aspect observations is further exploited in a probabilistic manner to accelerate the stereo-reconstruction process. To verify the effectiveness and reliability of the proposed method, 3D point cloud generation is tested on Ku-band drone SAR data for a domestic airport area.

Multi-Aspect Convolutional-Transformer Network for SAR Automatic Target Recognition

In recent years, synthetic aperture radar (SAR) automatic target recognition (ATR) has been widely used in both military and civilian fields. Due to the sensitivity of SAR images to the observation azimuth, the multi-aspect SAR image sequence contains more information for recognition than a single-aspect one. Nowadays, multi-aspect SAR target recognition methods mainly use recurrent neural networks (RNN), which rely on the order between images and thus suffer from information loss. At the same time, the training of the deep learning model also requires a lot of training data, but multi-aspect SAR images are expensive to obtain. Therefore, this paper proposes a multi-aspect SAR recognition method based on self-attention, which is used to find the correlation between the semantic information of images. Simultaneously, in order to improve the anti-noise ability of the proposed method and reduce the dependence on a large amount of data, the convolutional autoencoder (CAE) used to pretrain the feature extraction part of the method is designed. The experimental results using the MSTAR dataset show that the proposed multi-aspect SAR target recognition method is superior in various working conditions, performs well with few samples and also has a strong ability of anti-noise.

Automatic Building Detection for Multi-Aspect SAR Images Based on the Variation Features

Multi-aspect synthetic aperture radar (SAR) images contain more information available for automatic target recognition (ATR) than images from a single view. However, the sensitivity to aspect angles also makes it hard to extract and integrate information from multi-aspect images. In this paper, we propose a novel method based on the variations features to realize automatic building detection in the image level. First, to get a comprehensive description of target variation patterns, statistical characteristic variances are derived from three representative and complementary categories. Then, these obtained features are fused and put in the K-means classifier for prescreening, whose results are used as the training sets in supervised classification later to avoid manual labeling. Second, for more precise detection performance, finer features in vector forms are obtained by principal component analysis (PCA). The variation patterns of these feature vectors are explored in two different manners of correlation and fluctuation analyses and processed by separate support vector machines (SVMs) after fusion. Finally, the independent SVM detection results are fused according to a maximum probability rule. Experiments conducted on two different airborne data sets demonstrate the robustness and effectiveness of the proposed method, in spite of significant target signature variabilities and cluttered background.

Multi-Aspect SAR Target Recognition Based on Prototypical Network with a Small Number of Training Samples.

At present, synthetic aperture radar (SAR) automatic target recognition (ATR) has been deeply researched and widely used in military and civilian fields. SAR images are very sensitive to the azimuth aspect of the imaging geomety; the same target at different aspects differs greatly. Thus, the multi-aspect SAR image sequence contains more information for classification and recognition, which requires the reliable and robust multi-aspect target recognition method. Nowadays, SAR target recognition methods are mostly based on deep learning. However, the SAR dataset is usually expensive to obtain, especially for a certain target. It is difficult to obtain enough samples for deep learning model training. This paper proposes a multi-aspect SAR target recognition method based on a prototypical network. Furthermore, methods such as multi-task learning and multi-level feature fusion are also introduced to enhance the recognition accuracy under the case of a small number of training samples. The experiments by using the MSTAR dataset have proven that the recognition accuracy of our method can be close to the accruacy level by all samples and our method can be applied to other feather extraction models to deal with small sample learning problems.

A Man-Made Target Extraction Method Based on Scattering Characteristics Using Multiaspect SAR Data

Multiaspect synthetic aperture radar (SAR) can obtain the response of the target to the radar illumination in the angle dimension. The radar cross section data referring to scattering behavior can be obtained by multiaspect SAR. The anisotropy characteristics of the man-made target in the imaging scene can reflect its scattering changing behavior. Meanwhile, the amplitude characteristics of man-made target in some observing angles is target's another scattering property. Based on man-made target's anisotropy characteristics and amplitude characteristics, a dual-channel model (DCM) based on distribution and membership for man-made target extraction is generated in this article. The model considers man-made target's data changing in different observing aspects and strong points in specific aspects. First, the preprocessing of multiaspect data using low-rank matrix decomposition is discussed. Then, the parameters of SAR amplitude images’ distribution function are calculated by using expectation maximization method. Third, one channel is fuzzy C-means (FCM) method combining spatial neighborhood information for amplitude characteristics extraction. Another channel is statistical distribution model for anisotropy characteristics extraction. Finally, The calculated membership degree (MD) and statistical probability describe the man-made target and natural target. C-band circular SAR data is used to validate our method. The result of our DCM is compared with the result of only using a single channel model. The man-made targets are extracted better using the DCM.

DEM Generation With a Scale Factor Using Multi-Aspect SAR Imagery Applying Radargrammetry

Digital elevation model (DEM) generation using multi-aspect synthetic aperture radar (SAR) imagery applying radargrammetry has become a hotspot. The traditional radargrammetric method is to solve the rigorous radar projection equations to obtain the three dimensional coordinates of targets. In this paper, we propose a new DEM generation method based on the offset between multi-aspect images formed on ground plane. The ground object will be projected to different positions from different viewing aspect angles if the height of object is not equal to the height of imaging plane. The linear relationship between the offset of imaging positions and height of the object is derived and scale factor is obtained finally. Height information can be retrieved from offset of imaging positions directly through the DEM extraction model presented in this paper. Thus the solution to nonlinear equations point by point can be avoided. Real C band airborne circular SAR images is used to verify the proposed approach. When extracted DEM applied in multi-aspect imaging process, superimposition of multi-aspect images will no longer be defocusing and can achieve finer observation of the scanned scene.

Multi-aspect SAR target recognition based on space-fixed and space-varying scattering feature joint learning

ABSTRACTThe emerging synthetic aperture radar (SAR) multi-aspect imaging modes such as circular SAR (CSAR) bring new vitality and challenges to SAR imagery interpretation. In addition to the space-fixed features obtained from a single SAR image, multi-aspect SAR images can be used to extract the space-varying scattering features, which are beneficial to target recognition. However, due to the complexity of backscattering, the extraction of space-varying features has not yet been effectively explored. In this letter, a multi-aspect SAR target recognition framework based on a residual network (ResNet) and bidirectional long short-term memory (BiLSTM) network is proposed to address this. As a basis, the ResNet is used to extract the space-fixed scattering features from each single-aspect image, which is supervised by a joint loss. Then, the BiLSTM is employed to further extract the space-varying scattering features among the adjacent multi-aspect images through the recurrent learning. The experimental results show that the proposed method can achieve 100% accuracy for 10-class recognition, and 98.97% accuracy in case of a large depression angle. Compared with existing methods, it can implement high recognition accuracy and generalization with relatively simple combined neural networks.

Estimation and Removing of Anisotropic Scattering for Multiaspect Polarimetric SAR Image

Multiaspect Synthetic Aperture Radar (SAR) can generate high resolution images and target scattering signatures in different azimuth angles from the coherent integration of all subaperture images. However, mixed anisotropic scatters limit the application of traditional imaging theory. Anisotropic scattering may introduce errors in polarimetric parameters by decreasing the reliability of terrain classification and detection of variability. Thus a method is proposed for estimating and removing anisotropic scattering in multiaspect polarimetric SAR images. The proposed algorithm is based on the maximum likelihood and likelihood-ratio tests for the two-class case, while considering the speckle effect, the mechanism of removing the anisotropic scattering, and the monotonicity of the Constant False Alarm Rate (CFAR) detection function. We compare the polarimetric entropy before and after removing the anisotropic subapertures, and then validate the algorithm's potential in retrieving the target signature using a P-band quad-pol airborne SAR with circular trajectory.

Maximum-likelihood estimation for multi-aspect multi-baseline SAR interferometry of urban areas

The reconstruction of digital surface models (DSMs) of urban areas from interferometric synthetic aperture radar (SAR) data is a challenging task. In particular the SAR inherent layover and shadowing effects need to be coped with by sophisticated processing strategies. In this paper, a maximum-likelihood estimation procedure for the reconstruction of DSMs from multi-aspect multi-baseline InSAR imagery is proposed. In this framework, redundant as well as contradicting observations are exploited in a statistically optimal way. The presented method, which is especially suited for single-pass SAR interferometers, is examined using test data consisting of experimental airborne millimeterwave SAR imagery. The achievable accuracy is evaluated by comparison to LiDAR-derived reference data. It is shown that the proposed estimation procedure performs better than a comparable non-statistical reconstruction method.

基于压缩感知的多角度合成孔径雷达成像

Multi-aspect synthetic aperture radar (SAR) imaging is an important technique for synthesizing multi-aspect SAR data or information. Achieving multi-aspect SAR imagery is invaluable for resolution improvement, radar object visualization, classification and automatic object recognition (ATR). However, achieving multi-aspect SAR imaging is challenging because of the diversity of space sampling positions and the transmission of signal waveforms. In this study, a new method, based on the model of a multi-aspect SAR data collection, is proposed to achieve multi-aspect SAR imaging using compressive sensing and sparse reconstruction theory. First, the factors which dominate the performance of the measurement matrix were studied. It was found that the distribution of space sampling and the carrier frequency of the transmission signal were the key factors dominating the mutual coherence (MC) of the measurement matrix. The object sampling space was also a key factor in determining the robustness and resolution of multi-aspect SAR imaging. Second, the measurement matrix for the multi-aspect SAR was constructed, depending on previous analysis, by careful selection of the parameters and sampling of the object. Third, the stage-wise orthogonal matching pursuit (StOMP) algorithm was applied to achieve multi-aspect SAR imaging. This algorithm was effective in solving high-dimensional ill-posed equations. Extensive experiments validated the feasibility and stability of the proposed algorithm.

TARGET RECOGNITION FOR MULTI-ASPECT SAR IMAGES WITH FUSION STRATEGIES

Two fusion strategies for target recognition using multi- aspect synthetic aperture radar (SAR) images are presented for recognizing ground vehicles in MSTAR database. Due to radar cross- section variability, the ability to discriminate between targets varies greatly with target aspect. Multi-aspect images of a given target are used to support recognition. In this paper, two fusion strategies for target recognition using multi-aspect SAR images are proposed, which are data fusion strategy and decision fusion strategy. The recognition performance sensitivity to the number of images and the aspect separations is analyzed for those two target recognition strategies. The two strategies are also compared with each other in probability of correct classiflcation and operating e-ciency. The experimental results indicate that if we have a small number of multi-aspect images of a target and the aspect separations between those images are proper, the probability of correct classiflcation obtained by the two proposed strategies can be advanced signiflcantly compared with that obtained by the method using single image.

POLARIMETRIC SCATTERING MODELING AND INFORMATION RETRIEVAL OF SAR REMOTE SENSING --- A REVIEW OF FDU WORK

Synthetic aperture radar (SAR) imagery technology is one of most important advances in space-borne microwave remote sensing during recent decades. Completely polarimetric scattering from complex terrain surfaces can be measured. Polarimetric SAR (POLSAR), and its relevant technologies, such as POL-interferometric SAR (POLINSAR), bistatic SAR (POL-BISAR), high resolution (in m and dm resolution) SAR, inverse SAR (ISAR) etc., have been providing rich all-weather, all-time and high-resolution data and images of active miscrowave remote sensing. Fully understanding and retrieving information from polarimetric scattering signatures of natural media and SAR images have become a key issue for the SAR remote sensing and its broad applications. Many researches on polarimetric scattering and SAR imagery technology have been carried out (e.g., [1–6]). This paper presents a review of the research works in Fudan University (FDU) during recent years on theoretical modeling of the terrain surface for polarimetric scattering simulation and Mueller matrix solution, mono-static and bistatic SAR image simulation, new parameters for unsupervised surface classification, DEM inversion, change detection from multi-temporal SAR images, and reconstructions of buildings from multi-aspect SAR images, etc. [7–46]. Some applications are briefly reported. Corresponding author: Y.-Q. Jin (yqjin@fudan.ac.cn).

Persistence modeling of angularly dependent synthetic aperture radar imagery

An image persistence framework is developed to analyze azimuthally varying synthetic aperture radar (SAR) data. The model focuses on cases containing rich aspect data from a single depression angle. The goal is to replace the data's intrinsic viewing geometry dependencies with target-specific dependencies. Both direct mapping functions and cost functions are presented for data transformation. An intensity-only mapping function is realized to illustrate the persistence model in terms of a canonical example, visualization, and classification. The limitations of an intensity-only mapping function are also discussed. Because the new image-space output from the persistence model is closely tied to the radio frequency (RF) characteristics of two different targets and not to the collection geometry, it shows promise for integration into various automated target recognition (ATR) algorithms. Target classification potential was tested using the MSTAR database and simplified template matching schemes. The intensity-only transformation function permits a very low number of persistence images to represent the general RF target characteristics of an assortment of vehicles. The ability to represent target characteristics with relatively low resources could also be beneficial in ATR applications. Overall, the persistence framework shows strong potential as a new tool that can be used in the analysis of multiaspect SAR images.

Automatic Reconstruction of Building Objects From Multiaspect Meter-Resolution SAR Images

Reconstruction of 3-D objects from multiaspect high- resolution synthetic aperture radar (SAR) images is of great importance for SAR technology applications. In this paper, simple building objects are modeled as cuboids, and an approach for automatic reconstruction of 3-D building objects from multiaspect SAR images in meter resolution is developed. The edge detector of constant false alarm rate and a Hough transform technique for parallel line segment pairs are first employed to extract the parallelogram-like image of the building walls in SAR images. A set of probability density functions is presented to describe the object images and their multiaspect coherence. The maximum-likelihood estimation of an object is then derived from its multiaspect object images. A hybrid priority criterion is defined to evaluate the reliability of the reconstruction result. An automatic reconstruction algorithm is further developed to match object images of different aspects and, finally, to reconstruct the building objects. Besides, an iterative method is proposed for the coregistration of multiaspect building images. Four-aspect simulated images of a virtual scene and four-aspect Pi-SAR images over the campus of Tohoku University, Japan, are investigated. Reconstruction of building objects from their multiaspect images shows the fidelity of the whole process chain and the feasibility of 3-D objects automatic reconstruction from multiaspect SAR images. At last, a practical application that is based on spaceborne meter-resolution SAR is proposed.