A new method for color image segmentation based on FSVM

Author

Li, Lei ; Li, Jin-yan ; Ding, Wen-yan

Author_Institution

Sch. of Autom., Nanjing Univ. of Posts & Telecommun., Nanjing, China

Volume

2

fYear

2010

fDate

11-14 July 2010

Firstpage

664

Lastpage

668

Abstract

Image segmentation is very essential and critical to image processing and pattern recognition. It is known that, color image segmentation approaches are based on monochrome segmentation approaches operating in different color spaces. So in this paper, an improved method which uses the FSVM (fuzzy support vector machines) algorithm for color image segmentation in the HSI (hue-saturation-intensity) color space was proposed. Experiments show that the proposed algorithm has better effect and lower computational complexity on color image segmentation than other method.

Keywords

fuzzy logic; image colour analysis; image recognition; image segmentation; support vector machines; FSVM algorithm; HSI color space; color image segmentation; computational complexity; fuzzy support vector machines; hue-saturation-intensity; image processing; monochrome segmentation; pattern recognition; Color; Cybernetics; Image color analysis; Image segmentation; Machine learning; Support vector machines; Training; Color image segmentation; Color space; FSVM (fuzzy support vector machines); HSI (hue-saturation-intensity);

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics (ICMLC), 2010 International Conference on

Conference_Location

Qingdao

Print_ISBN

978-1-4244-6526-2

Type

conf

DOI

10.1109/ICMLC.2010.5580556

Filename

5580556