A Novel Template-Based Change Detection Technique for Harbor Ship Target

A novel template-based change detection technique for harbor ship target has been proposed. Unlike classical methods, the proposed approach exploits information by the assistance of template map. Firstly, high resolution optical images are used to make binary templates. Secondly, the template and the two images to be detected are registered separately, through which we obtain the region of interesting by calculating the spatial transformation of the template and the image sequence. Thirdly, we obtain the segmentation sea area images through weighted fusion of template and different temporal images, then use Constant False Alarm Rate (CFAR) detection algorithm to extract the changed targets. Finally, target features are extracted to perform change analysis. Experimental results confirm the effectiveness of the proposed technique.