Unsupervised Texture Segmentation Based on Image Decomposition

Abstract

In this paper, we propose a new image decomposition model devoted to texture segmentation. By coupling some constraint terms in the model, the given textured image is decomposed into a sum of two functions with different basic characteristics. Then the textural feature space is segmented by using level set and active contours technology. Compared with some recent approaches, our method is more convincing and works well for synthetic textured images and ones in the real world