Tensor based image segmentation

In recent years spectral clustering has become on e of the most popular clustering algorithms. It is a simple yet powerful method for finding structure in data using spectral properties of an associated pairwise similarity matrix. It is simple to implement, can be solved efficiently by standard linear algebra software, and very often outperforms traditional clustering algorithms such as the k-means algorithm. In this paper we propose a new way of image segmentation based on specifically created similarity matrix and based on it, very simple segmentation algorithm. The algorithm is theoretically motivated and demonstrated on nontrivial examples.