A novel robust and fast segmentation of the color images using fuzzy classification c-means

Author

Toure, Mohamed Lamine ; Beiji, Zou ; Musau, Felix

Author_Institution

Sch. of Inf. Sci. & Eng., Central South Univ., Changsha, China

Volume

4

fYear

2010

fDate

22-24 June 2010

Abstract

This paper brings out a method for segmentation of color images based on fuzzy classification. It proceeds in a first step by a fine segmentation using the algorithm of fuzzy c-means (FCM). The method then applies a test fusion of fuzzy classes. The result is a coarse segmentation, where each region is the union of elementary regions grown from FCM. The fuzzy C-Means (FCM) clustering is an iterative partitioning method that produces optimal c-partitions, the standard FCM algorithm takes a long time to partition a large data set. The proposed FCM program must read the entire data set into a memory for processing. Our results show that the system performance is robust to different types of images.

Keywords

image colour analysis; image segmentation; pattern clustering; color image segmentation; fuzzy c-means clustering; fuzzy classification c-means; iterative partitioning method; optimal c-partitions; Clustering algorithms; Color; Fuzzy sets; Image segmentation; Iterative algorithms; Iterative methods; Partitioning algorithms; Robustness; System performance; Testing; Classification; FCM; FuzzyLogic; Merge regions; Optimal c-partitions; Segmentation;

fLanguage

English

Publisher

ieee

Conference_Titel

Education Technology and Computer (ICETC), 2010 2nd International Conference on

Conference_Location

Shanghai

Print_ISBN

978-1-4244-6367-1

Type

conf

DOI

10.1109/ICETC.2010.5529667

Filename

5529667