A fast CFAR detection algorithm based on the G0 distribution for SAR images

The statistical model of clutter is a key factor which determines the performance of a CFAR algorithm for target detection in SAR image. The distribution is able to accurately model the homogeneous, heterogeneous and extremely heterogeneous regions in a single look or multi-look SAR image. But its applicability is greatly limited by its disadvantages that the parameter estimation is complex and the threshold cannot be acquired easily. In view of these problems, this paper uses the moment estimation and dichotomy method to complete the parameter estimation and the threshold acquirement. In addition, pre-filtering the target candidate regions and iterative calculation are used to increase the efficiency. A new algorithm is proposed, aiming at the effectiveness and efficiency at the same time. The experimental results prove its practicability.