Radar automatic target recognition (RATR) technology is fundamental but complicated system engineering that combines sensor, target, environment, and signal processing technology, etc. It plays a significant role in improving the level and capabilities of military and civilian automation. Although RATR has been successfully applied in some aspects, the complete theoretical system has not been established. At present, deep learning algorithms have received a lot of attention and have emerged as potential and feasible solutions in RATR. This paper mainly reviews related articles published between 2010 and 2022, which corresponds to the period when deep learning methods were introduced into RATR research. In this paper, the current research status of radar target characteristics is summarized, including motion, micro-motion, one-dimensional, and two-dimensional characteristics, etc. This paper reviews the progress of deep learning methods in the feature extraction and recognition of radar target characteristics in recent years, including space, air, ground, sea-surface targets, etc. Due to more and more attention and research results published in the past few years, it is hoped that this review can provide potential guidance for future research and application of deep learning in fields related to RATR.

Compared with the commonly used lidar and visual sensors, the millimeter-wave radar has all-day and all-weather performance advantages and more stable performance in the face of different scenarios. However, using the millimeter-wave radar as the Simultaneous Localization and Mapping (SLAM) sensor is also associated with other problems, such as small data volume, more outliers, and low precision, which reduce the accuracy of SLAM localization and mapping. This paper proposes a millimeter-wave radar SLAM assisted by the Radar Cross Section (RCS) feature of the target and Inertial Measurement Unit (IMU). Using IMU to combine continuous radar scanning point clouds into “Multi-scan,” the problem of small data volume is solved. The Density-based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm is used to filter outliers from radar data. In the clustering, the RCS feature of the target is considered, and the Mahalanobis distance is used to measure the similarity of the radar data. At the same time, in order to alleviate the problem of the lower accuracy of SLAM positioning due to the low precision of millimeter-wave radar data, an improved Correlative Scan Matching (CSM) method is proposed in this paper, which matches the radar point cloud with the local submap of the global grid map. It is a “Scan to Map” point cloud matching method, which achieves the tight coupling of localization and mapping. In this paper, three groups of actual data are collected to verify the proposed method in part and in general. Based on the comparison of the experimental results, it is proved that the proposed millimeter-wave radar SLAM assisted by the RCS feature of the target and IMU has better accuracy and robustness in the face of different scenarios.

The Beijing-Tianjin Intercity Railway (BTIR) is an important transportation hub connecting Beijing and Tianjin. It is the China's first high-speed railway reaching a speed of 350 km/h and has been in operation for ten years. Since the foundation of the BTIR is made up of soft and loose soil, hence suffers greatly from potential deformation. In order to investigate the impact of ground subsidence on the railway line, 26 TerraSAR-X SpotLight images, 23 TerraSAR-X StripMap images, and 31 Sentinel-1 Interferometric Wide images were processed to obtain the deformation of the Yizhuang areas along the BTIR from January 2015 to March 2017 by permanent scatterers interferometric synthetic aperture radar technology. The deformation results of these three datasets are self-consistent. Through the vertical and horizontal deformation of the railway using fusion of the railway deformation results from different line of sight direction, we found that the deformation of the railway line is mainly in the vertical direction. Furthermore, the high precision spatial-temporal evolution of the deformation along the BTIR is revealed. Finally, the advantages and limitations of different SAR resolution are analyzed and summarized.

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.

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 $\mathrm{Ku}$-band drone SAR data for a domestic airport area.