The Maximum Variance Between Clusters Method of Image Segmentation Based on Particle Swarm Optimization

Author

Li, Jian-ming ; Chi, Zhong-Xian ; Yu, Li-qiang ; Zhang, Feng ; Jiang, Qiao-qiao

Author_Institution

Dept. of Comput. Sci., Dalian Univ. of Technol.

fYear

2006

fDate

13-16 Aug. 2006

Firstpage

3765

Lastpage

3769

Abstract

This essay proposes a maximum variance between clusters method of image segmentation (OTSU) based on PSO. The method in this paper makes use of particle swarm algorithm and achieves a great acceleration to the traditional OTSU. On that basis, we also applied the parallelism technology in particle-swarm algorithm and find an optimal threshold, so we can segment images with this threshold. The result proves that we not only raised the speed highly but also achieved a great efficiency, due to the discrete global searching algorithm we adopted

Keywords

image segmentation; particle swarm optimisation; pattern clustering; search problems; cluster method; discrete global searching algorithm; image segmentation; maximum variance; parallelism technology; particle swarm optimization; Acceleration; Computer science; Cybernetics; Gray-scale; Image analysis; Image processing; Image recognition; Image segmentation; Machine learning; Parallel processing; Particle swarm optimization; Robustness; Image segmentation; OTSU; PSO;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2006 International Conference on

Conference_Location

Dalian, China

Print_ISBN

1-4244-0061-9

Type

conf

DOI

10.1109/ICMLC.2006.258680

Filename

4028726