Binary Tree-Based Clustering Algorithm and Used in Color Image Segmentation

Abstract

An effective clustering algorithm based on binary tree is proposed in this paper. A roughly extraction of the color image is gotten by constructing the self-adapting binary tree. A C_means clustering algorithm is designed based on the extraction. This algorithm is to improve the segmentation's accuracy of the binary tree's leaves. Experiments have approved that this new method can be implemented efficiently. Experiments show the results of high segmentation accuracy and mass information in origin color image at the same time.