High-resolution SAR Image Target Detection And Feature Extraction

Synthetic aperture radar?SAR?has been widely used in both military and civil fields by virtue of its imaging characteristics.Target detection and feature extraction in SAR images is an important part of SAR Automatic Target Recognition?SAR-ATR?system and it is of great significance for target recognition and classification in SAR images.However,with the resolution of the acquired SAR image becoming higher and higher,the detection and feature extraction of the target of interest in SAR images also encounter challenges and problems.Therefore,the research work of target detection and feature extraction in high-resolution SAR images is carried out in this paper.The main work and contributions are summmarized as follows:1.Combining the algorithms based on the CFAR and morphological processing detection algorithm to detect the target in high-resolution SAR images.Firstly,the CFAR detection principle and other relevant theoretical knowledge are introduced in detail.Then the target detection in SAR images based on morphological algorithm is introduced.Finally,the two detection algorithms are combined to realize the target detection in high-resolution SAR images.Moreover,the visual and friendly Matlab Graphical User Interface based on the target detection in SAR images has been designed.2.The frequently-used geometric feature,corner feature and edge feature in high-resolution SAR images are extracted.Harris corner detection algorithm is used to extract the corner features in high-resolution SAR images,and the edge detection algorithm is used to extract the edge feature in the SAR images.The different target features in several kinds of different SAR images are extracted and the difference between them is analyzed.Moreover,the difference of target corner and edge feature in the same one SAR image before and after morphological processing is compared and analyzed.3.A target detection algorithm based on information theory and Harris corner detection in SAR images is proposed.Firstly,the SAR image is pretreated,and next,it is divided into superpixel patches by using the improved SLIC superpixel generation algorithm.Then the self-information of the superpixel patches is calculated and the threshold T₁ is set to select the candidate superpixel patches.And then the extended neighborhood weighted information entropy growth rate threshold T₂ is set to eliminate false alarm candidate superpixel patches.Finally,the Harris corner detection algorithm is used to process the detection result and the corner number threshold T₃ is set to filter out the false alarm patches,and the final SAR image target detection result is obtained.The simulation experiments and result analyses of target detection in high-resolution SAR image based on the proposed information theory and Harris corner detection algorithm are carried out.The simulation objects include the SAR images of marine ships with land and SAR images of marine ships without land,and include the SAR images of land tank and vehicles.Make a comparation and use the detection algorithm to detect the targets in high-resolution SAR images based on CFAR combining with morphological processing algorithm and other detection algorithms.The effectiveness and superiority of the proposed algorithm is verified by the simulation experiments.