Computer Engineering and Applications ›› 2013, Vol. 49 ›› Issue (13): 149-151.

Previous Articles     Next Articles

Image segmentation based on minimum spanning tree

LI Ying, DAI Fang, HAO Yong, ZUO Tao   

  1. School of Science, Xi’an University of Technology, Xi’an 710054, China
  • Online:2013-07-01 Published:2013-06-28

基于最小生成树的图像分割

黎  莹,戴  芳,郝  勇,左  涛   

  1. 西安理工大学 理学院 应用数学系,西安 710054

Abstract: Based on minimum spanning tree, a method of image segmentation using improved minimum spanning tree is presented. The procedure of generating minimum spanning tree is reduced, and then the similarity-based neighborhood graph method is applied to merge image split by minimum spanning tree. The proposed method saves the time of image segmentation, meanwhile, based on the effective merger the better segmentation results are obtained.

Key words: minimum spanning tree, NNG(Nearest Neighbor Graph) method, image segmentation

摘要: 基于最小生成树思想,给出了一种利用改进的最小生成树进行图像分割的方案,减少了最小生成树的构建过程,对初分割的结果利用NNG算法进行合并。该方案节约了分割时间,并且对分割后的图像进行了有效的合并,达到了较好的分割效果。

关键词: 最小生成树, 相似邻近图, 图像分割