Spectral clustering based unsupervised change detection in SAR images

An unsupervised change detection method based on spectral clustering and difference image methods for multitemporal single-channel single-polarization synthetic aperture radar (SAR) images is proposed. The difference image is generated by integrating the typical difference image method with Non-Local Filter, which exploits both the spatial neighborhood information and gray similarity information, and can well reduce the speckle noises of SAR images. The spectral clustering algorithm is employed to cluster the difference image into two clusters and get the change map. Compared with traditional clustering algorithms, such as k-means, SC can recognize the clusters of unusual shapes and obtain the globally optimal solutions. Experimental results confirm the effectiveness of the proposed techniques.