Image Segmentation by Unsupervised Sparse Clustering

In this paper, we present a novel solution of image segmentation based on positiveness by regarding the segmentation as one of the graph-theoretic clustering problems. On the contrary to spectral clustering methods using eigenvectors, the proposed method tries to find an additive combination of positive components from an originally positive data-driven matrix. By using the positiveness constraint, we obtain sparsely clustered results which are closely related to human perception and thus we call this method sparse clustering. The proposed method adopts a binary tree structure and solves a model selection problem by automatically determining the number of clusters using intra- and intercluster measures. We tested our method with various kinds of data such as points, gray-scale, color, and texture images. Experimental results show that the proposed method provides very successful and encouraging segmentations.

• Type: Conference
• Year: 2005
• Where: WACV
• Authors: Byoung-Ki Jeon, Yun-Beom Jung, Ki-Sang Hong
• Keywords: Graph-theoretic Clustering Problems, Positive Data-driven Matrix, Spectral Clustering Methods, WACV 2005