It is a feasible and promising way to utilize deep neural networks to learn and extract valuable features from synthetic aperture radar (SAR) images for SAR automatic target recognition (ATR). However, it is too difficult to effectively train the deep neural networks with limited raw SAR images. In this paper, we propose a new approach to do SAR ATR, in which a multiview deep learning framework was employed. Based on the multiview SAR ATR pattern, we first present a flexible mean to generate adequate multiview SAR data, which can guarantee a large amount of inputs for network training without needing many raw SAR images. Then, a unique deep convolutional neural network containing a parallel network topology with multiple inputs is adopted. The features of input SAR images from different views will be learned by the proposed network layer by layer; meanwhile, the learned features from the distinct views are fused in different layers progressively. Therefore, the proposed framework is able to achieve a superior recognition performance, and requires only a small number of raw SAR images for network training samples generation. Experimental results have shown the superiority of the proposed framework based on the Moving and Stationary Target Acquisition and Recognition data set.

Real-aperture radar can be utilized to realize forward-looking imaging by antenna scanning the imaging region. However, low azimuth resolution seriously affects its practical application. Although traditional super-resolution methods could enhance azimuth resolution to a certain extent, effective preservation of contour information for important targets still remains to be a problem. In this article, a method of total variation-sparse (TV-sparse) multiconstraint deconvolution is proposed to improve azimuth resolution of forward-looking imaging as well as preserve contour information of important targets. Since our interested targets usually appear to be sparse, the sparse constraint of the target is introduced first to achieve high resolution of forward-looking images, which may cause the loss of target contour information in the meantime. Second, total variation (TV) constraint is introduced based on the sparse constraint, converting traditional single-constraint super-resolution problem to a multiconstraint problem. We then use the split Bregman algorithm (SBA) to solve the multiconstraint problem, whose solution is the super-resolution image of radar forward-looking region. Compared with traditional super-resolution methods, the proposed method can improve the azimuth resolution of radar forward-looking imaging as well as better restore target contour information by adjusting respective weights of sparse constraint and TV constraint. Finally, the performance of the proposed method is validated with the simulation and measured data.

Synthetic aperture radar (SAR) is a microwave remote sensing system. It has a broad scope of applications in both military and civilian fields. Benefited from the latest advances in deep learning, SAR automatic target recognition technology has made an excellent breakthrough However, most existing methods ignore the large variation of scattering characteristics of SAR target images with different azimuths, which limits the performance and practical application. The SAR images under different azimuths contain distinct feature information, and the images under adjacent azimuths are correlated in terms of features. Therefore, extracting the feature information of images under adjacent azimuths and leveraging their correlation can improve the recognition performance. In this article, we proposed a multiview attention convolutional neural network with long short-term memory (LSTM) network to extract and fuse the features from images with adjacent azimuths. It adopts multiple convolutional modules to extract deep features from each single-view SAR image and spatial attention module to locate the information of the target and suppress the useless noise. Then, the LSTM module performs feature fusion based on the correlation of features obtained from adjacent azimuths. Finally, based on these multiview images, deep features are extracted and fused to obtain precise recognition results. Experiments are performed on the moving and stationary target acquisition and recognition dataset, and the results have verified the effectiveness of the proposed method.

Maritime surveillance is indispensable for civilian fields, including national maritime safeguarding, channel monitoring, and so on, in which synthetic aperture radar (SAR) ship target recognition is a crucial research field. The core problem to realizing accurate SAR ship target recognition is the large inner-class variance and inter-class overlap of SAR ship features, which limits the recognition performance. Most existing methods plainly extract multi-scale features of the network and utilize equally each feature scale in the classification stage. However, the shallow multi-scale features are not discriminative enough, and each scale feature is not equally effective for recognition. These factors lead to the limitation of recognition performance. Therefore, we proposed a SAR ship recognition method via multi-scale feature attention and adaptive-weighted classifier to enhance features in each scale, and adaptively choose the effective feature scale for accurate recognition. We first construct an in-network feature pyramid to extract multi-scale features from SAR ship images. Then, the multi-scale feature attention can extract and enhance the principal components from the multi-scale features with more inner-class compactness and inter-class separability. Finally, the adaptive weighted classifier chooses the effective feature scales in the feature pyramid to achieve the final precise recognition. Through experiments and comparisons under OpenSARship data set, the proposed method is validated to achieve state-of-the-art performance for SAR ship recognition.

Convolutional neural networks (CNNs) have dominated the synthetic aperture radar (SAR) automatic target recognition (ATR) for years. However, under limited SAR images, the width and depth of the CNN-based models are limited, and the widening of the received field for global features in images is hindered, which finally leads to the low performance of recognition. To address these challenges, we propose a convolutional transformer (ConvT) for SAR ATR few-shot learning (FSL). The proposed method focuses on constructing a hierarchical feature representation and capturing global dependencies of local features in each layer, named global in local. A novel hybrid loss is proposed to interpret few SAR images in the forms of recognition labels and contrastive image pairs, construct abundant anchor-positive and anchor-negative image pairs in one batch, and provide sufficient loss for the optimization of ConvT to overcome few sample effect. An auto augmentation is proposed to enhance and enrich the diversity and amount of few training samples to explore the hidden feature in few SAR images and avoid over-fitting in SAR ATR FSL. Experiments conducted on the moving and stationary target acquisition and recognition (MSTAR) dataset have shown the effectiveness of our proposed ConvT for SAR ATR FSL. Different from the existing SAR ATR FSL methods using additional training datasets, our method achieved pioneering performance without other SAR target images in training.

Marine surveillance radar is widely used in marine monitoring for its ability of observing sea surface all-time and all-weather. However, the radar target detection performance is seriously affected by the existence of sea clutter. In this article, we propose a new sea clutter suppression method based on machine learning approach. We first employ a cyclic structure network with a pair of generative adversarial networks to sufficiently learn the characteristics of sea clutter, which converts the problem of sea clutter suppression as a transformation from the clutter radar data domain to the clutter-free radar data domain. In addition, we propose a target-consistency loss for the cost function of the network to effectively preserve the target information while suppressing the sea clutter. Therefore, the proposed method can not only effectively remove the sea clutter from the radar data but also protect the target information from being damaged during sea clutter suppression, thereby achieving excellent sea clutter suppression performance. Experimental results have shown the superiorities of the proposed sea clutter suppression method on both simulated and measured marine surveillance radar data.