基于图论和FCM的图像分割算法

吴秋红; 吴谨; 朱磊; 李娟

doi:10.3788/YJYXS20163101.0112

您当前的位置：

首页 >

文章列表页 >

基于图论和FCM的图像分割算法

显示与图像 | 更新时间：2020-08-12

- 基于图论和FCM的图像分割算法
- Image segmentation algorithm based on graph theory and FCM
- 液晶与显示 2016年31卷第1期页码：112-116
- 作者机构：
  
  武汉科技大学信息科学与工程学院,湖北武汉,430081
- 作者简介：
- 基金信息：
- DOI：10.3788/YJYXS20163101.0112
  中图分类号： TP391.41
- 收稿日期：2015-03-23，
  
  修回日期：2015-04-22，
  
  纸质出版日期：2016-01-05
- 稿件说明：
移动端阅览
吴秋红, 吴谨, 朱磊等. 基于图论和FCM的图像分割算法[J]. 液晶与显示, 2016,31(1): 112-116

WU Qiu-hong, WU Jin, ZHU Lei etc. Image segmentation algorithm based on graph theory and FCM[J]. , 2016,31(1): 112-116
吴秋红, 吴谨, 朱磊等. 基于图论和FCM的图像分割算法[J]. 液晶与显示, 2016,31(1): 112-116 DOI： 10.3788/YJYXS20163101.0112.

WU Qiu-hong, WU Jin, ZHU Lei etc. Image segmentation algorithm based on graph theory and FCM[J]. , 2016,31(1): 112-116 DOI： 10.3788/YJYXS20163101.0112.

摘要

针对FCM算法缺少空间关联信息且计算量大的问题

本文提出一种结合图论和FCM的图像分割算法。首先

引入图论算法对图像进行预处理

将图分割为子图。接着

对分割后的子图进行FCM聚类得到聚类中心。然后

提出一种基于聚类中心颜色和空间信息的加权距离

作为并查集算法的合并准则。最后

采用改进的并查集算法对聚类结果进行区域合并。实验结果表明

本文算法在保证图像分割质量的同时提高了图像分割速度。

Abstract

Aiming at fuzzy C-means clustering algorithm lacking spatial information and large amount of calculation

an algorithm combined graph theory with fuzzy C-means clustering is presented in this paper. Firstly

the graph is divided into subgraph by graph theory algorithm. After that

clustering center of the subgraph is obtained by FCM algorithm. And then

a weighted distance based on the color of clustering center and space information is proposed as the merging criterion of union-find sets algorithm. Finally

the improved union-find sets is used to merge the regions of clustering results. The experimental results indicate that the proposed algorithm can ensure the quality of image segmentation with better performance.

关键词

Keywords

references

DUNN J C. A fuzzy relative of the ISODATA process and its use in detecting compact well-separated clusters[J]. Journal of Cybernetics, 1973, 3(3):32-57.

BEZDEK J C. A convergence theorem for the fuzzy ISODATA clustering algorithms[J].IEEE Transactions on pattern Analysis and Machine Intelligence, 1980, PAMI-2(1):1-8.

GONG M G, LIANG Y, SHI J, et al. Fuzzy c-means clustering with local information and kernel metric for image segmentation[J].IEEE Transactions on Image Processing, 2013, 22(2):573-584.

李云松,冯玉东,张国锋.基于快速模糊C均值聚类的图像粗集分割[J].兰州理工大学学报,2013,39(1):92-96. LI Y S, FENG Y D, ZHANG F. Image rough sets segmentation based on fast fuzzy C-means clustering[J].Journal of Lanzhou University of Technology, 2013, 39(1):92-96. (in Chinese)

郭海涛,刘丽媛,赵亚鑫,等.基于MAR与FCM聚类的声呐图像分割[J].仪器仪表学报,2013,34(10):2322-2327. GUO H T, LIU L Y, ZHAO Y X,et al. Sonar image segmentation based on MAR and FCM clustering[J]. Chinese Journal of Scientific Instrument, 2013, 34(10):2322-2327. (in Chinese)

周晓明,李钊,刘雄英.一种基于改进FCM的自动图像分割算法[J].华南理工大学学报(自然科学版),2014,42(3):1-7. ZHOU X M, LI Z, LIU X Y. An automatic image segmentation algorithm based on Improved FCM[J].Journal of South China University of Technology (Natural Science Edition), 2014, 42(3):1-7. (in Chinese)

龚劬,廖武忠,卢力,等.基于图论的快速FCM图像分割算法[J].计算机工程,2012,38(8):192-194,197. GONG Q, LIAO W Z, LU L,et al. Fast FCM image segmentation algorithm based on graph theory[J]. Computer Engineering, 2012, 38(8):192-194, 197. (in Chinese)

辛月兰,汪西莉.基于图割与改进模糊C均值的图像分割[J].计算机应用与软件,2014,31(7)206-209,293. XIN Y L, WANG X L. Image segmentation based on graph cuts and improved fuzzy C-means[J].Computer Applications and Software, 2014, 31(7):206-209, 293. (in Chinese)

李旭超,刘海宽,王飞,等.图像分割中的模糊聚类方法[J].中国图象图形学报,2012,17(4):447-458. LI X C, LIU H K, WANG F,et al. The survey of fuzzy clustering method for image segmentation[J]. Journal of Image and Graphics, 2012, 17(4):447-458. (in Chinese)

林选伟,吴谨.基于交互式图论的目标边缘检测算法[J].计算机工程与科学,2014,36(8):1571-1575. LIN X W, WU J. A novel target edge detection algorithm based on interactive graph theory[J].Computer Engineering and Science, 2014, 36(8):1571-1575. (in Chinese)

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于边缘增强和特征融合的伪装目标分割

基于图像分割的点云去噪方法

基于NODE-UNet++和标记分水岭算法的红细胞图像分割

自适应动态肤色检测技术研究进展

基于Faster R-CNN的仪表识别方法