UNSUPERVISED TARGET DETECTION IN SAR IMAGES USING SCATTERING CENTER MODEL AND MEAN SHIFT CLUSTERING ALGORITHM

A new framework for ship detection in synthetic aperture radar (SAR) images is proposed. We focus on the task of locating reflective small regions using scattering centers model and clustering algorithm. Unlike most of the approaches in ship detection, we address an algorithm that incorporates SAR super-resolution imaging and mean shift clustering instead of parameter estimation. Our approach is validated by a series of tests on real SAR images and compared with other ship detection algorithms, demonstrating that it configures a novel and effcient method for ship-detection purpose.