Minimum Spanning Tree and Color Image Segmentation
Xuexi Zhang, Yimin Yang · 2008
Image segmentation based on graph theory is mainly used for gray image now, and thresholding of segmentation should be predefined. Combining with maximum between -and- within -class in statistics theory, this paper suggests an unsupervised method for color image segmentation. The image is mapped into an weighted undirected graph, the pixels are considered as nodes, and minimum spanning tree is constructed by Kruskal algorithm .The best thresholding is obtained by maximum objective function to realize unsupervised segmentation. Experiment results show that the new algorithm ensures the color image segmentation excellent disturbance attenuation performance and better separability.