Change detection in optical remote sensing images based on watershed segmentation

A novel approach is presented for change detection of very high resolution (VHR) remote sensing images in this paper, which is achieved by an improved watershed segmentation and mutual correlation. Specifically, the input multitemporal images are watershed segmented separately in the first stage. And then fuse the segmentation regions of two segmented images, which is also a combination of object outlines in other words. Finally, the change mask can be acquired by the mutual correlation method. Experimental results show the superiority of the proposed method over the Markov random filed (MRF) method to the VHR urban optical remote sensing images.