The research on island change detection techniques of multiple-band oriented high resolution remote sensing image

Digital change detection is the computerized process of identifying changes in the state of an object, or other earthsurface features, between different data. During the last years, a large number of change detection methods have evolved that differ widely in refinement, robustness and complexity. Some traditional change detection methods could not any more adapt to high resolution remote sensing images. The prime tendency of remote sensing change detection is from pixels level to object level. In the paper, with respect to the views of object-oriented change detection in remote sensing images, an unsupervised technique for change detection (CD) in very high geometrical resolution images is proposed, which is based on the use of morphological filters. This technique integrates the nonlinear and adaptive properties of the morphological filters with a change vector analysis (CVA) procedure. Different morphological operators are analyzed and compared with respect to the CD problem. Alternating sequential filters by reconstruction proved to be the most effective, permitting the preservation of the geometrical information of the structures in the scene while filtering the homogeneous areas. We collect two multi-temporal SPOT5 remote sensing images to analyze YangSan island change detection in this procedure as above mentioned. Experimental results confirm the effectiveness of the proposed technique. It increases the accuracy of the CD in high remote sensing change detection as compared with the standard CVA approach.