Detection of circular oil tanks based on the fusion of SAR and optical images

This paper aims to detect circular oil tanks automatically by using both SAR and optical images. The proposed method consists of two main steps: first, the bright areas are extracted from the SAR image to obtain the candidate regions for oil tanks; second, shape detection is performed in the optical image on these corresponding candidate areas, before we finally determine whether they are oil tanks or not. Prior knowledge, such as the validate dimensions of oil tanks and the group emerging property, is taken into account in this method. The experimental results demonstrate the effectiveness of the proposed method.