Sparse representation-based spectral clustering for SAR image segmentation

Abstract

A new method, sparse representation based spectral clustering (SC) with Nystr&#246;m method, is proposed for synthetic aperture radar (SAR) image segmentation. Different from the conventional SC, this proposed technique is developed by using the sparse coefficients which obtained by solving <i>l</i>1 minimization problem to construct the affinity matrix and the Nystr&#246;m method is applied to alleviate the segmentation process. The advantage of our proposed method is that we do not need to select the scaling parameter in the Gaussian kernel function artificially. We apply the proposed method, k-means and the classic spectral clustering algorithm with Nystr&#246;m method to SAR image segmentation. The results show that compared with the other two methods, the proposed method can obtain much better segmentation results.