A study of automation recognition of bridge in SAR image

According to river and bridge features in SAR (Synthetic Aperture Radar, SAR) image as well as the position relationship between bridge and river, the automatic recognition method of bridge is presented. First, the gradient of an image is calculated by gradient operator in order to get the gradient image, and river region is segmented by Fisher criterion. Then, a doubtful bridge region is extracted by morphology method, and segmented by Fisher criterion. Finally, the bridge is recognized according to the features. Experimental results show that this method can automatically and rapidly recognize bridge.