An experiment on MSTAR data for CFAR detection in lognormal and weibull distributed sar clutter

In this paper, a constant false alarm rate (CFAR) target detection algorithm is presented. Here, lognormal and Weibull distributions are considered for modeling clutter amplitude of synthetic aperture radar(SAR) images. Then Kullback-Leibler (KL) goodness-of-fit test is employed to evaluate the histogram matching accuracy of estimated probability density functions (pdfs) of selected models with SAR clutter data histogram. With the use of appropriate background clutter model, CFAR detection algorithm is applied to moving and stationary target acquisition and recognition (MSTAR) data. Experimental results demonstrate that, accuracy of CFAR detector is more for Weibull clutter compared to the same for lognormal clutter.