Passive ground camouflage target recognition based on gray feature and texture feature in SAR images

Abstract

Recently, Synthetic Aperture Radar plays a significant role in military reconnaissance and battle damage assessment. However, comparing with optic image, SAR image interpretation is complex because of their imaging characteristics. Furthermore, it is more difficult to distinguish camouflage targets in SAR image. In this paper, we proposed a new method to recognize real targets and camouflage targets. On the basis of SAR images, we studied the difference between real targets and passive camouflage targets at first. Then, we extract gray feature and texture feature of all the targets in an image. Finally, targets were divided into two clusters by using K-means algorithm. The simulation result demonstrates feasibility of passive ground camouflage target recognition.