A fast target detection framework for SAR imagery

In this paper, we propose a fast target detection framework for synthetic aperture radar (SAR) imagery. The crucial issue of this method is the potential target area map which is obtained from a preliminary screening. With reference to the potential target area map, we can conduct local constant false alarm rate (CFAR) detection only in the potential target areas as well as eliminate the potential target pixels from the estimation of the local clutter distribution. The experiment results show that the proposed fast target detection framework not only has high detection accuracy but also has a great improvement on efficiency.