Detecting vehicle target in SAR image with high moment

Vehicle detection in SAR image is attractive field. High moments included numerous information is proposed to exploit in the field. It not only restrains Gaussian noise automatically, but also suppresses non-gaussian noise. Man-made targets in are different from the background clutter. Moreover, the influence of the shadow about man-made targets is reduced in the algorithm of high moments. We utilize high moments to avoid to analyst the complex scatter theory. By analyzing the relevent theory and practical calculated method of high order moments, we compare the difference of probability of detection between the target chip and clutter chip to achieve the goal of detection. Through processing lots of actual SAR data which is added different kinds of noise, we compare the performance between high order moments and low-order moments method.