Automatic recognition of ISAR Images: Target shapes features extraction

Automatic target recognition (ATR) is a challenging task for several computer vision applications. It requires efficient, accurate, and robust methods for target detection and target identification. Deep learning has shown great success in many computer vision applications involving color RGB images. However, the performance of these networks in ATR with infrared sensor data needs further investigation. In this paper, we propose a multistage automatic target detection and recognition (ATDR) system that performs both target detection and target classification on infrared (IR) imagery using deep learning. Our system processes large IR image frames where targets take <1 % of the total number of pixels. First, we train a state-of-the-art object detector you only look once (YOLO) to localize all potential targets in the input image frame. Then, we train a convolutional neural network (CNN) to identify these detections as targets or false alarms. In this second phase, we adapt and analyze the performance of three CNN architectures: a compact and fully connected CNN, VGG16 with batch normalization, and a wide residual neural network (WRN). We also explore the use of a loss function that optimizes directly the area under the receiver operating characteristic (ROC) curve (AUC), and adapt it to our ATR application. To enhance the robustness of the proposed ATR to perturbation and variations introduced during the detection stage, we train our CNN classifiers on automatically detected targets using YOLO, in addition to ground truth bounding boxes and apply selected data augmentation techniques. To simulate real testing environments, where the spatial location of the targets within the image frame is unknown, only YOLO-detected boxes are used during validation. We evaluate our ATDR on a real benchmark dataset that includes different vehicles captured at different resolutions. Our experiments have shown that YOLO can detect most of the targets at the expense of generating a high number of false alarms. We show that the VGG-16 network with batch normalization, which is the best performing model, can correctly identify the classes of the targets, as well as classify the majority of YOLO’s false detections into an additional nontarget class. We also show that the proposed training modification to optimize an AUC-based loss function for ATR proved to be advantageous mainly in identifying difficult targets.

Radar target recognition is an important research direction in the field of radar information processing. Low resolution radar ground target classification and recognition still has broad application prospects in modern military and civil fields. Convolutional neural network (CNN) has attracted more and more attention in the field of radar automatic target recognition, because it does not need Feature Engineering and has superior classification performance. The optimization algorithm of neural network is proposed, and the effects of different network structures on the recognition effect are compared, in this paper. The results show that the multi-scale wide residual network multi-target classification algorithm based on optimized auto-encoder has better recognition effect in low resolution radar target recognition.

Radar data can be presented in various forms, unlike visible data. In the field of radar target recognition, most current work involves point cloud data due to computing limitations, but this form of data lacks useful information. This paper proposes a semantic segmentation network to process high-dimensional data and enable automatic radar target recognition. Rather than relying on point cloud data, which is common in current radar automatic target recognition algorithms, the paper suggests using a radar heat map of high-dimensional data to increase the efficiency of radar data use. The radar heat map provides more complete information than point cloud data, leading to more accurate classification results. Additionally, this paper proposes a dimension collapse module based on a vision transformer for feature extraction between two modules with dimension differences during dimension changes in high-dimensional data. This module is easily extendable to other networks with high-dimensional data collapse requirements. The network's performance is verified using a real radar dataset, showing that the radar semantic segmentation network based on a vision transformer has better performance and fewer parameters compared to segmentation networks that use other dimensional collapse methods.

DARPA/Air Force Research Laboratory Moving and Stationary Target Acquisition and Recognition (MSTAR) program is developing state-of-the-art model based vision approach to Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR). The model-based approach requires using off-line developed target models in an on-line hypothesize-and-test manner to compare predicted target signatures with image data and output target reports. Central to this model-based ATR is the PEMS (Predict-Extract-Match-Search) subsystem. The Search module is critical to PEMS by providing intelligent control to traverse the hypothesis feature space. A major MSTAR goal is to demonstrate robust ATR for variations in targets including partially hidden targets. This paper will provide an update on the technology being developed under MSTAR and the status of this model based ATR research, specifically concentrating on the Search Module.

A synthetic aperture radar (SAR) automatic target recognition can effectively improve the utilization efficiency of SAR image data. In order to improve the generalization and accuracy of SAR image target recognition, a convolutional neural network model is proposed to be applied to SAR image target recognition. Firstly, the network structure is designed for the characteristics of SAR image. Secondly, the introduction of dense block network structure improves the generalization performance of the model. Finally, Dropout reduced the computational complexity of the network and improved the generalization performance of the model. Experimental data were obtained from the United States Moving and Stationary Target Acquisition and recognition database. Experimental results of 10 types of target recognition showed that the overall recognition rate of the improved convolutional neural network is 99.18%. The convolutional neural network model proposed in this study improves the accuracy and generalization of the network, and is an effective method for target recognition of SAR images.

In the field of synthetic aperture radar (SAR) automatic target recognition (ATR), deep learning-driven methods perform well when numerous training samples are available. However, in practical SAR application scenarios, the number of available training samples is often scarce due to the limitation of observation conditions, which leads to major challenges for deep learning-driven methods in real-world deployment. Currently, the problem of ATR with only a few training samples has attracted much attention in the SAR field. In this article, we put forward an ATR method called azimuth-guided feature embedding network with dual inference mechanism (AGFEN-DIM) to achieve fewshot SAR target recognition. To be specific, the feature extraction model, i.e. azimuth-guided feature embedding network (AGFEN) is composed of the azimuth embedding module (AEM) and the dynamic feature embedding module (DFEN). Among them, AEM resorts to a set of azimuth-parameterized convolution kernels, aiming to make full use of information relevant to the current radar imaging environment. The proposed DFEN is capable of dynamically extracting the discriminative features of the target according to the input image under the guidance of azimuth-parameterized convolutional kernels. The dual inference mechanism (DIM) effectively integrates the inference results of improved prototype classification and label propagation strategy to boost the reliability and robustness of the recognition model for few-shot SAR target recognition tasks. Extensive experiments on the moving and stationary target acquisition and recognition (MSTAR) dataset demonstrate that the proposed AGFEN-DIM surpasses many state-of-the-art SAR ATR methods in various application scenarios.

Combination of global and local filters for robust SAR target recognition under various extended operating conditions

This paper describes the US Army's efforts under the ATR relational template matching program to develop a multi-sensor forward-looking IR (FLIR) and millimter wave (MMW) radar automatic target recognition (ATR) algorithm. The general problem consists of identifying ground targets at low depression angles of less than 6 degrees and ranges of 500 to 6000 m. However, the algorithm is directly applicable to other target sets, including air targets, and ground targets at high depression angles. Our phase I goal was to provide a proof- of-principle demonstration of this new model-based methodology. We did this on simulated FLIR imagery for a three target class problem. A comprehensive test and evaluation was performed at the Night Vision and Electronic Sensors Directorate. Our goal in phase II was to demonstrate this new algorithm approach on realistic field-collected second-generation FLIR imagery with a larger target set consisting of ten ground vehicles. With phase III, our goal was to demonstrate multisensor fusion on FLIR (first generation) fused with laser radar. This was demonstrated on a four-ground-target-class problem. The limitations to first-generation FLIR data and four targets was due purely to restrictions of the field-collected data. Our goal in phase IV is to demonstrate this methodology for FLIR fused with MMW radar data. Phase IV will be a three-year effort, with the first year concentrating on using the MMW radar data for detection, reduction in false alarms, and the provision of accurate range (i.e., scale) to the FLIR algorithm. The goal of the second and third years will be to fuse the MMW data into not only the detection phase of the algorithm but also the recognition and identification portions of the algorithm. This paper presents the general model-based methodology used for the FLIR and the neural network methodology used to provide the FLIR with the MMW detections and scale. Results are provided for the FLIR-only algorithm for short range and moderate range scenarios given an accurate estimate of target range.

In recent years, convolutional neural network (CNN) has been increasingly considered as a promising technology for military and homeland security applications. The fusion of CNN and Support vector machine (SVM), a popular traditional machine learning approach, has received intensive attention in the field of synthetic aperture radar (SAR) automatic target recognition (ATR). This paper, firstly, discusses the effects of some preprocessing and image enhancement methods on the performance of SAR ATR, starting with the pre-trained AlexNet in a transfer-learning based approach. Secondly, the architecture of AlexNet is modified to form a new model suitable for SAR ATR. Finally, we propose a hybrid model associated with the success of the learning feature of our CNN model and the ability of SVM to process high-dimensional dataset effectively. To evaluate the proposed method, experiments are performed on the Moving and Stationary Target Acquisition and Recognition (MSTAR) public database. The comparative results demonstrate that these preprocessing and enhancement methods prior to the deep-learning process are not necessary since the feature representation ability of AlexNet is already powerful. Furthermore, experimental results on the benchmark MSTAR dataset illustrate the effectiveness of the proposed new model. On classification of ten-class targets, the commonly used translation augmentation for training data has been performed. By combining the CNN and SVM, the classification accuracy percentages can be slightly improved for our proposed new model.

The existing deep learning studies on synthetic aperture radar (SAR) automatic target recognition (ATR) mainly focus on the utilization of the amplitude of SAR image via convolutional neural network (CNN), while the electromagnetic scattering information is rarely considered. Given that scattering centers (SCs) can characterize the target&#x2019;s electromagnetic scattering characteristics and describe the target&#x2019;s physical structure information, the SC feature should be helpful for SAR ATR. Therefore, we propose a novel SAR ATR method that combines electromagnetic scattering information and graph convolutional network (GCN) effectively and directly. Specifically, we model each extracted SC as a node to convert the SCs into graph data. The constructed graph is learned via GCN to describe the target&#x2019;s physical structure information, where the features of different GCN layers are fused to avoid the over-smoothing of GCN. Label smoothing is combined with GCN for the first time to alleviate the overfitting caused by the limited training data. To the best of our knowledge, this study is the first to introduce the GCN for effectively utilizing the SCs, proving that the structural characteristics of the SCs of SAR targets are remarkably beneficial for recognition. Extensive experimental results on the measured moving and stationary target acquisition and recognition (MSTAR) dataset show that our method can obtain superior recognition performance compared with the existing methods.

The existing synthetic aperture radar (SAR) automatic target recognition (ATR) methods have shown impressive results in static scenarios, yet the performance drops sharply as new categories of targets are continuously increased. In response to this problem, this letter proposes a novel ATR method named multi-level adaptive knowledge distillation network (MLAKDN) to achieve incremental SAR target recognition. To be specific, an adaptive weighted distillation strategy is first proposed, which can alleviate the model from forgetting the knowledge of old categories by distilling multi-stage soft label information of old categories at the classification level. Then, a feature distillation method based on gradient maximum criterion is developed to filter and distill discriminative features, so as to further recall more knowledge of old categories at the feature level. Meanwhile, a model rebalancing technique is designed to effectively strike the balance of the model on new categories and old categories. Finally, a weighted incremental classification loss is presented to train the whole model. Experiments on the moving and stationary target acquisition and recognition (MSTAR) dataset and the synthetic and measured paired labeled experiment (SAMPLE) dataset illustrate that the proposed method is superior to some state-of-the-arts for incremental SAR target recognition tasks.

In the radar automatic target recognition (RATR) system using high resolution range profile (HRRP), if a test sample has not been trained in training phase, it would lead to a full miss classification in test phase. In this paper, we design a classifier based on generalized confidence, which can efficiently refuse-recognize a new target. Firstly, principal component analysis (PCA) method is used to extract feature vector from every sample. Secondly, the classifier calculates each feature vector's generalized confidence on mahalanobis distance. Consequently, the distribution of confidence is used to define a refuse-recognition threshold for each training target. In test phase, for each trained-target in the database, we suppose that the test sample belongs to current target, calculate the generalized confidence, judge whether the test sample really belongs to the target or not via comparing the confidence with the target's refuse-recognition threshold. The final class is determined by vote. The experimental results demonstrate the effectiveness of the proposed algorithms.

This study extends the theory of the micro-Doppler effect into the multistatic domain, and considers how the multistatic micro-Doppler signature (µ-DS) will affect radar automatic target recognition (ATR). Real multistatic µ-DS of personnel targets are examined and their nature compared with theory and simulated results. It is demonstrated that the use of multistatic µ-DSs would increase the robustness of radar ATR systems to the problem of self-occlusion, where the target obscures itself. Redundancy in the multistatic µ-DS is identified, and how this may reduce the data fusion demands of multi-aspect classifiers is discussed. Also, information theory is used to demonstrate that the multistatic µ-DS contains more target information than the monostatic case. This result leads to the prediction that multi-aspect µ-DS-based radar ATR would have improved performance compared to single-aspect solutions.

Radar target automatic recognition is a comprehension problem. We have researched the practical radar data and concluded some valuable viewpoint. With it we put forward a whole scheme to achieve radar automatic target recognition. First the OS CAFR and binary accumulation algorithm is taken to accurately detect target. Here we lower self-adapt threshold to get more targets which maybe be false target. Then we use target trace management to remove the illusion target. The novel algorithm, called as endpoint detection based on moving window, can get more valuable data to classification. For 3D target we must build more templets in classification. So we use Neural Net as classification algorithm to reduce memory space. The simulation results prove that this scheme can get better recognition results.

This paper considers the use of the software package WinBUGS for an application to a previously documented technique for automatic radar target recognition. Mixture models under a Bayesian formalism are used to model the probability densities of radar measurements from given classes. WinBUGS is then used to construct a Markov chain Monte Carlo sampler that allows implementation of this Bayesian approach. The model developed is applied to radar range profiles from seven classes of military ships.