作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2012, Vol. 38 ›› Issue (18): 194-197. doi: 10.3969/j.issn.1000-3428.2012.18.052

• 图形图像处理 • 上一篇    下一篇

一种基于图论的图像分割算法

张 乾1,2,冯夫健1,林 鑫1,王 林1,2   

  1. (1. 贵州省模式识别与智能系统重点实验室,贵阳 550025;2. 贵州民族大学教务处,贵阳 550025)
  • 收稿日期:2011-11-21 修回日期:2012-01-15 出版日期:2012-09-20 发布日期:2012-09-18
  • 作者简介:张 乾(1984-),男,讲师、硕士研究生,主研方向:图像处理,模式识别;冯夫健、林 鑫,硕士研究生;王 林,教授、博士
  • 基金资助:
    国家自然科学基金资助项目(60965001);贵州省科学技术基金资助项目(黔科合J字[2011]2207号);贵州省科学技术基金委员会-贵州民族学院联合基金资助项目(黔科合J字LKM[2011]04号)

An Image Segmentation Algorithm Based on Graph Theory

ZHANG Qian1,2, FENG Fu-jian1, LIN Xin1, WANG Lin1,2   

  1. (1. Key Laboratory of Pattern Recognition and Intelligent Systems of Guizhou Province, Guiyang 550025, China;2. Academic Affairs Office, Guizhou Minzu University, Guiyang 550025, China)
  • Received:2011-11-21 Revised:2012-01-15 Online:2012-09-20 Published:2012-09-18

摘要: 针对图像分割应用中阈值难以确定的问题,提出一种基于图论的图像分割算法。利用二维高斯分布函数给出边权重函数的动态自适应系数,结合区域间、区域内的相似度函数定义差距函数,得到适合区域合并的动态判定函数。实验结果表明,与其他算法相比,该算法的图像分割效果较好,花费时间较少。

关键词: 图像分割, 图论, 判定函数, 最短路径, 结构相似度

Abstract: It is difficult to determine the threshold and weight coefficients in image segmentation. Aiming at this problem, this paper proposes a new image segmentation algorithm based on graph theory. It uses two-dimensional Gaussian distribution function as the edge weight coefficient for it is dynamic adaptive, as well as resorts the regional-regional and inter-regional similarity function to delimit the gap function, and finds a dynamics determine function for the area to be merged or segmented. Experimental results show that this algorithm is better than other similar ones in image segmentation, and it can save the running time.

Key words: image segmentation, graph theory, determine function, the shortest path, structural similarity

中图分类号: