A comprehensive and well-structured review on the application of deep learning (DL) based algorithms, such as convolutional neural networks (CNN) and long-short term memory (LSTM), in radar signal processing is given. The following DL application areas are covered: i) radar waveform and antenna array design; ii) passive or low probability of interception (LPI) radar waveform recognition; iii) automatic target recognition (ATR) based on high range resolution profiles (HRRPs), Doppler signatures, and synthetic aperture radar (SAR) images; and iv) radar jamming/clutter recognition and suppression. Although DL is unanimously praised as the ultimate solution to many bottleneck problems in most of existing works on similar topics, both the positive and the negative sides of stories about DL are checked in this work. Specifically, two limiting factors of the real-life performance of deep neural networks (DNNs), limited training samples and adversarial examples, are thoroughly examined. By investigating the relationship between the DL-based algorithms proposed in various papers and linking them together to form a full picture, this work serves as a valuable source for researchers who are seeking potential research opportunities in this promising research field.

The traditional inverse synthetic aperture radar (ISAR) imaging uses the range-Doppler (RD) type of methods. The compressive sensing (CS)-based ISAR imaging is capable of obtaining good target images of high contrast and less sidelobe with much less downsampling data. However, the real application of CS ISAR imaging is limited by the time-consuming iteration-based image reconstruction. The image quality is also limited by the performance of sparse representation of the target scene. In recent years, deep learning methods, more specifically the convolutional neural network (CNN), has shown its capability in signal recovery with downsampling or noncomplete data. The well-trained CNN can extract high-level abstract feature representation from the input data autonomously and exploit it in the signal recovery. We are interested in exploiting the CNN to enhance the CS ISAR imaging capability. The successful training of CNN always requires many thousand annotated training samples. This limits the application of CNN to the radar imaging field where large amount of training data cannot be obtained as easy as in other fields, e.g., computer vision. We propose a fully CNN (FCNN) for ISAR imaging. The constructed FCNN has a multistage decomposition and multichannel filtering architecture and has no fully connected layers. It can work with very few training samples as compared to existing CNN-based imaging networks. The imaging results of real ISAR data show that the proposed FCNN-based ISAR imaging method outperforms the state-of-the-art CS ISAR imaging methods in both image quality and computational efficiency.

Mass production of high-quality synthetic SAR training imagery is essential for boosting the performance of deep-learning (DL)-based SAR automatic target recognition (ATR) algorithms in an open-world environment. To address this problem, we exploit both the widely used Moving and Stationary Target Acquisition and Recognition (MSTAR) SAR dataset and the Synthetic and Measured Paired Labeled Experiment (SAMPLE) dataset, which consists of selected samples from the MSTAR dataset and their computer-generated synthetic counterparts. A series of data augmentation experiments are carried out. First, the sparsity of the scattering centers of the targets is exploited for new target pose synthesis. Additionally, training data with various clutter backgrounds are synthesized via clutter transfer, so that the neural networks are better prepared to cope with background changes in the test samples. To effectively augment the synthetic SAR imagery in the SAMPLE dataset, a novel contrast-based data augmentation technique is proposed. To improve the robustness of neural networks against out-of-distribution (OOD) samples, the SAR images of ground military vehicles collected by the self-developed MiniSAR system are used as the training data for the adversarial outlier exposure procedure. Simulation results show that the proposed data augmentation methods are effective in improving both the target classification accuracy and the OOD detection performance. The purpose of this work is to establish the foundation for large-scale, open-field implementation of DL-based SAR-ATR systems, which is not only of great value in the sense of theoretical research, but is also potentially meaningful in the aspect of military application.

The registration of synthetic aperture radar (SAR) and optical images is critical in multimodal remote sensing image fusion. In recent years, deep learning-based registration networks have been continuously introduced. However, owing to the significant disparities in viewing angles and radiometric properties between SAR and optical images, current deep learning methods struggle to fully exploit the physical properties of radar imaging. Additionally, many existing matching networks typically perform only a forward pass, resulting in suboptimal model performance. This paper proposes a cyclic iterative registration SAR mechanism network (termed as CIRSM-Net) for the registration of SAR and optical images. Firstly, we design a learning module that integrates the radar equation with a microwave scattering model to capture deep features from SAR images, and design a corresponding scattering feature loss to aid in better generalization across various radar images. Then, to explore optimization methods for matching networks, this study proposes a strategy of multiple iterative optimizations within the matching network. Specifically, it integrates speeding-up radiation-variation insensitive feature transform (RIFT2) supervision in the backend matching network and iteratively optimizes the final output. Finally, during the iteration process, we propose an innovative matching loss function that combines the rotation invariance supervision of RIFT2 with iterative optimization techniques to enhance feature matching accuracy. Experimental results on both public and our own datasets additionally confirm the effectiveness and superiority of the proposed approach, demonstrating its significant potential for practical applications.

Pansharpening refers to the fusion between a multispectral (MS) image with abundant spectral information and a panchromatic (PAN) image with high spatial resolution to obtain a high spatial resolution multispectral (HRMS) image. The traditional pansharpening methods often ignore the effect of path-radiation caused by scattering from different atmospheric components, and the few methods that introduce haze correction only calibrate each band of the MS image individually, without exploring the intrinsic correlation among different bands. To address this problem, low rank tensor completion pansharpening based on haze correction (LRTCP) is proposed. The haze-line prior is first introduced into the joint haze correction of MS and PAN images, and obtain the pre-modulated images with the help of the improved high-pass modulation (HPM) injection scheme. We then use tensor completion to simulate the degradation problem by applying low-tubal-rank tensor complementation to the process of reconstructing HRMS images, thus constructing a low rank tensor completion pansharpening model based on haze correction. Finally, the alternating direction multiplier (ADMM) is employed to find the solution of the proposed approach, producing the final fusion result. Comprehensive qualitative and quantitative assessment of reduced- and full-resolution datasets from different satellites shows that the proposed method outperforms the state-of-the-art methods.

Ground armored weapons have a high detection value in military operations. Satellite synthetic aperture radar (SAR) cannot accurately detect military targets with meter-level sizes limited by resolution of sensors. Airborne SAR have strict experimental conditions and cannot be applied in actual battlefield environments. MiniSAR sensors, which combine the advantages of submeter-level ultrahigh resolutions and flexible flight, play a crucial role in recognizing military targets. In this paper, various small military targets in real complex ground scenarios are detected with the MiniSAR of NUAA. However, there are still two difficulties. First, because of a limitation in the number of flight circles, the number of obtainable military target samples is not sufficient to adapt to the traditional deep learning methods that rely on a large number of image samples. Second, due to the imaging systems and different depression angle of MiniSAR, the SAR images of MiniSAR suffer from the same deformation challenge as the moving and stationary target acquisition and recognition (Mstar) with high depression angle. To address these two challenges, we propose a FASAR-Net framework based on few-shot learning with meta learning and adversarial domain learning, combined with the inherent scattering features of the SAR targets. Furthermore, we validate the reliability and accuracy of this algorithm on Mstar and our datasets, and the result of recognizing small SAR targets is compared with our algorithm and other classical algorithms. We conclude that the proposed algorithm has high accuracy in the recognition of the deformation small targets under the few sample condition.