A mover detection method based on 2-D fuzzy entropy in SAR images

In this paper, a new technique for the moving targets detection combining incoherent and coherent information of the Synthetic Aperture Radar (SAR) images is proposed based on the measure of a two-dimensional (2-D) fuzzy entropy principle. Firstly, it extracts incoherent and coherent information from images in pixel and builds a 2-D histogram. Secondly, according to 2-D fuzzy function, image pixels are fuzzy classified into two groups: changed and unchanged then 2-D fuzzy entropy is calculated. Finally, Differential Evolution (DE) algorithm is used to maximize the entropy and the optimal parameters of the fuzzy function are obtained, which procures effective moving target detection results.