SAR Image Target Research Articles

Application of Gaussian process model in SAR image target recognition

Application of Gaussian process model in SAR image target recognition

SAR image target recognition method using based on Res-Net deep features

SAR image target recognition method using based on Res-Net deep features

Synthetic Aperture Radar SAR Image Target Recognition Algorithm Based on Attention Mechanism

SAR images contain a large amount of noise, and related algorithms will cause high complexity when increasing the accuracy. To overcome this problem, a neural network model based on the attention mechanism was proposed in this paper. The model extracted information in two stages. It gradually extracts high-level features by reducing noise first and then adding hybrid attention. First, use dual-channel one-dimensional convolution to reconstruct the residual shrinkage network to construct a lightweight and efficient feature module, which improved the information extraction of the module with the consumption of a small amount of computing resources. Then, it was used as the backbone for model construction. Subsequently, mixed adaptive pooling was adopted to improve the maximum pooling. After that, dimensionality was reduced by pooling and linear interpolation was used to increase dimensionality, so as to generate feature weights of mixed dimension. Tests were performed on MSTAR dataset. The results showed that compared with the advanced algorithms, the proposed model in this paper can greatly reduce the amount of parameters and complexity while ensuring accuracy. The robustness test demonstrated that the model can effectively identify images with noise being added.

SAR Image Target Recognition Based on Monogenic Signal and Sparse Representation

It is necessary to recognize the target in the situation of military battlefield monitoring and civilian real‐time monitoring. Sparse representation‐based SAR image target recognition method uses training samples or feature information to construct an overcomplete dictionary, which will inevitably affect the recognition speed. In this paper, a method based on monogenic signal and sparse representation is presented for SAR image target recognition. In this method, the extended maximum average correlation height filter is used to train the samples and generate the templates. The monogenic features of the templates are extracted to construct subdictionaries, and the subdictionaries are combined to construct a cascade dictionary. Sparse representation coefficients of the testing samples over the cascade dictionary are calculated by the orthogonal matching tracking algorithm, and recognition is realized according to the energy of the sparse coefficients and voting recognition. The experimental results suggest that the new approach has good results in terms of recognition accuracy and recognition time.

基于改进残差注意力网络的SAR图像目标识别

基于改进残差注意力网络的SAR图像目标识别

SSS-YOLO: towards more accurate detection for small ships in SAR image

ABSTRACT Aiming at the low detection rate and high false alarm in small ship detection in SAR images, we propose a small-scale ship detection algorithm based on convolutional neural network in this paper. First, we redesign the feature extraction network according to the characters of ship targets in SAR images. The modified network can enrich the spatial and semantics information of small ships. Then, we propose the Path Argumentation Fusion Network (PAFN) to improve the fusion of different feature maps. PAFN uses bottom-up and top-down ways to fuse more location information and semantic information. Both these two optimizations can enhance the detection for small ships. We evaluate our model based on the open SAR-Ship-Dataset and Gaofen-3 SAR images. The experiment results show that our method has excellent performance for small ship detection compared with other deep learning models. Our model improves AP by 6.5% and has higher detection efficiency compared with the baseline YOLOv3 model.

Velocity Estimation of Moving Targets in Multi-channel SAR images by Optimally-weighted Velocity Averaging of Multiple Pixels

ABSTRACT A scheme is proposed for the radial-velocity estimation of moving targets in multi-channel SAR systems. A moving target consists of multiple pixels in SAR images, generally. The radial velocity of each pixel can be estimated from its interferometric phase. The radial velocity of the moving target is then obtained by averaging the radial-velocity estimates of its multiple pixels. In the averaging, the weight of each radial-velocity estimate is designed to minimize the mean-square-error (MSE) between the estimate and the truth. This scheme has a higher estimation accuracy than those conventional schemes, which take the radial-velocity estimate of one pixel as the final radial-velocity estimate of the moving target. The experimental results of both simulated and real data show the effectiveness of this scheme.

结合多分辨率表示和复数域CNN的SAR图像目标识别方法

结合多分辨率表示和复数域CNN的SAR图像目标识别方法

MdpCaps-Csl for SAR Image Target Recognition With Limited Labeled Training Data

Although convolutional neural networks (CNN) have shown excellent performance in many image recognition tasks, it commonly requires a lot of labeled data, and the recognition effect is frequently unsatisfied due to the limited labeled training data. In recent years, capsule network (CapsNet) has been shown to achieve a high recognition accuracy with a small group of training samples. In this study, a class separable loss based on cosine similarity is suggested to enhance the distinguishability of the extracted network. It is added as a regularization term to the original loss function to train the network, narrowing the intra-class difference and increasing the inter-class difference in each iteration. Meanwhile, a multi-dimensional parallel capsule module is established to obtain robust features and spatial relationships from the original images. Feature maps from convolution of different levels are extracted as the input of this module. Structural features derived from low-level convolution and semantic features derived from high-level convolution are used for low-dimensional capsule coding and high-dimensional capsule coding, respectively. In our experiment, the general moving and stationary target acquisition and recognition (MSTAR) database is used. We find that the accuracy of the multi-dimensional parallel capsule network with class separable loss (MdpCaps-Csl) is 99.79% using all training samples, which is higher than most current recognition methods. More importantly, the accuracy is up to 97.73% even if only 10% training samples are applied, indicating MdpCaps-Csl can make excellent performance upon limited training samples.

Morphology-Combined Gradient Boosting for Recognizing Targets in SAR Images

Abstract This article proposes a novel method for recognizing objects in synthetic aperture radar images. The target is initially detected using a proposed morphology-based segmentation process and is further confirmed by classifying the objects. The identified target after the proposed segmentation process is subjected to feature extraction using Zernike moments, which efficiently downsamples the features and makes them rotationally invariant. The features are classified using a tree-based method called gradient boosting. Gradient boosting, by far, has shown very promising results on various kinds of data sets. The main focus is to bring to light this kind of tree-based architecture for target recognition in satellite imagery as well as to propose a framework that reduces the overall time and improves the efficiency of the process. It outperforms the previous state-of-the-art methods for this data size. The proposed method is compared with existing techniques to measure and evaluate its performance.

Stochastic configuration network‐based SAR image target classification approach

Stochastic configuration network‐based SAR image target classification approach

SAR image target‐detection method based on two‐stage CFAR detector with G 0 distribution

CFAR detection algorithm is widely used in the typical target detection of SAR images. In SAR image target detection, the simple statistical distribution models, such as Gaussian distribution, normal distribution, Weibull distribution etc., are obviously not suitable for the fitting of the background clutter distribution of missile-borne SAR images, and the G 0 distribution is suitable for extremely unstable regions and its expression is relatively simple. A two-stage CFAR detector with G 0 distribution is designed and the distribution of ground SAR images is statistically analysed by using the CA-CFAR detector. A target in a relatively non-homogenous background usually has a long tail, and the detection results of the two-stage CFAR detector based on the G0 distribution are much better.

Saliency detection of targets in polarimetric SAR images based on globally weighted perturbation filters

In this paper, a saliency detection for Polarimetric Synthetic Aperture Radar (PolSAR) images is proposed based on weighted perturbation filters. Auxiliary data is demanded to identify polarimetric vector of targets, for a canonical perturbation filter. Only if the target signature was available and accurate, it would be satisfiable to apply the filter in practice. Besides, not every target can usually be detected by an individual filter, because of variant polarimetric characteristics of targets with respect to different aspects or shapes. To overcome these drawbacks, several perturbation filters are combined in the proposed method. By initializing with different parameters, these filters decompose PolSAR data into their index maps. Then, aiming to find out filters of interest, i.e., ones related to target pixels, we assume that targets to detect are sparse in PolSAR image. Thus, saliency weights are assigned to the filters, based on Jaccard distances of their index maps. Therein, the spatial sparseness between objects and their surrounding derives high weights for corresponding filters. And then, after globally fusion of refined filtering responses with the weights, saliency map is generated for every local pattern in PolSAR image. Finally, the target regions are extracted from this map, by thresholding and morphological operation. Experiments performed on real and simulated PolSAR data verify the performance of this method, in comparison with several common PolSAR detectors. Also, the proposed method finds out most targets in ground truth, without auxiliary polarimetric information provided.

An Improved SAR Image Target Recognition Algorithm

Target recognition is a key step in the interpretation of SAR images. For existing SAR image recognition methods based on sparse representation, the recognition rate is still not high enough. Based on the analysis of the impact recognition rate, the target region and shadow region characteristics in SAR images are combined. In this paper, a SAR image target recognition method based on sparse representation and stretching is proposed. In this method, a new training sample image is generated by stretching the training sample image, a sparse dictionary is constructed by using the existing training sample image and a new training sample image, and the joint sparse representation of the target region and the shadow region is solved by the reconstruction. The minimum error criterion completes the SAR image target recognition. The target recognition method of this paper is tested by using MSTAR real-measured SAR image. The results show that the recognition rate of this method is higher than that of existing methods, which verifies the effectiveness of the proposed method.

An Improved Superpixel-Level CFAR Detection Method for Ship Targets in High-Resolution SAR Images

To achieve efficient ship detection in high-resolution synthetic aperture radar images, an improved superpixel-level constant false alarm rate (CFAR) detection method is proposed with three modifications. First, the weighted information entropy (WIE) describes the statistical characteristics of superpixels, yielding a better distinction between target and clutter superpixels. Second, a two-stage CFAR detection scheme is proposed to detect target superpixels, including global detection and local detection. Specifically, the WIE-based global detector is utilized to prescreen candidate target superpixels (CTSs) and then the local CFAR detector is adaptively conducted over the selected CTSs to refine target superpixels. Third, the superpixels selected by the global detection can be excluded as outliers for clutter estimation, thus the disturbance from adjacent targets in multitarget situations can be reduced and more accurate detection results can be obtained. Compared with CFAR methods implemented with the sliding window technique, the computational burden of the proposed method is significantly reduced without loss of detection performance. Experiments of ship detection on three TerraSAR-X datasets are presented to validate the effectiveness of the proposed method.

SAR Image Target Detection Method Based on Sparse Representation

SAR Image Target Detection Method Based on Sparse Representation

Deep kernel learning method for SAR image target recognition.

With the development of deep learning, research on image target recognition has made great progress in recent years. Remote sensing detection urgently requires target recognition for military, geographic, and other scientific research. This paper aims to solve the synthetic aperture radar image target recognition problem by combining deep and kernel learning. The model, which has a multilayer multiple kernel structure, is optimized layer by layer with the parameters of Support Vector Machine and a gradient descent algorithm. This new deep kernel learning method improves accuracy and achieves competitive recognition results compared with other learning methods.

Ship Targets Discrimination Algorithm in SAR Images Based on Hu Moment Feature and Texture Feature

To discriminate the ship targets in SAR images, this paper proposed the method based on combination of Hu moment feature and texture feature. Firstly,7 Hu moment features should be extracted, while gray level co-occurrence matrix is then used to extract the features of mean, variance, uniformity, energy, entropy, inertia moment, correlation and differences. Finally the k-neighbour classifier was used to analysis the 15 dimensional feature vectors. The experimental results show that the method of this paper has a good effect.

BMVT-M을 이용한 IR 및 SAR 융합기반 지상표적 탐지

Infrared (IR) target detection is one of the key technologies in Automatic Target Detection/Recognition (ATD/R) for military applications. However, IR sensors have limitations due to the weather sensitivity and atmospheric effects. In recent years, sensor information fusion study is an active research topic to overcome these limitations. SAR sensor is adopted to sensor fusion, because SAR is robust to various weather conditions. In this paper, a Boolean Map Visual Theory-Morphology (BMVT-M) method is proposed to detect targets in SAR and IR images. Moreover, we suggest the IR and SAR image registration and decision level fusion algorithm. The experimental results using OKTAL-SE synthetic images validate the feasibility of sensor fusion-based target detection.

New structure for multi-aspect SAR image target recognition with multi-level joint consideration

A new structure for SAR target recognition using multi-aspect SAR images is proposed. The new structure applies new approaches in the two levels of SAR image target recognition, which are data level and decision level, and combines the benefits of the two. Projections onto convex sets (POCS) super-resolution reconstruction algorithm is used in the data level of the new structure, which is to advance the resolution of the SAR image by using a series of multi-aspect SAR images. Weighted Bayes decision fusion algorithm is proposed in the decision level to jointly consider the benefit from the data level. All outcomes from the classifiers are fused in the decision level to generate the final result, which combines the multi-level benefits. Verification and analysis is performed to the proposed structure with multi-target image data in MSTAR database. Experimental results indicate that using the proposed structure for multi-aspect SAR images with multi-level joint consideration, the recognition rate is significantly advanced than that using single SAR target image. Meanwhile, the recognition rate by this structure is also higher than that using individual level approach.