Automatic circular oil tank detection in high-resolution optical image based on visual saliency and Hough transform

Abstract

It has great significance for the military and civilian application to detect oil tanks and locate oil depots automatically from remote sensing images. A novel and effective oil tank detection method is proposed by visual saliency and Hough transform in this paper. To separate oil tank regions from the background, we propose a new visual saliency model by mathematical morphology based on multi-scale histogram contrast. We do oil tank detection using Hough transform. The SVM algorithm is utilized to confirm that whether the circle is an oil tank. Experimental results show that this method can effectively detect oil tank on large-field, high-resolution optical images and achieved promising results in precise.