DocumentCode :
2634948
Title :
An Improved FCM algorithm for Color Image Segmentation
Author :
Kaiqi, Zou ; Zhiping, Wang ; Ming, Hu
Author_Institution :
Univ. Key Lab. of Inf. Sci. & Eng., Dalian Univ., Dalian
fYear :
2008
fDate :
18-20 June 2008
Firstpage :
200
Lastpage :
200
Abstract :
In this paper, an improved fuzzy C-means clustering (IFCM) algorithm for color image segmentation is proposed to solve the problem of heavy calculating burden and the disadvantage that clustering performance is affected by initial cluster centers for FCM, which is simple and easy to implement in color Image segmentation. For one thing, the quick subtractive clustering (QSC) is used for getting initial cluster centers of the image data points. For another, the first component of color feature set discovered by Ohta is chosen as the one-dimensional eigenvector. In order to reduce the computational complexity, the mapping from pixel space to eigenvector space is used for modifying the object function. Furthermore, combined the two problems of cluster centers initialization and cluster validity goes research to find optimizing the number of clusters. Experiments show that the proposed algorithm has better effect and lower computational complexity on color image segmentation.
Keywords :
eigenvalues and eigenfunctions; fuzzy set theory; image colour analysis; image segmentation; pattern clustering; color image segmentation; computational complexity; eigenvector; improved fuzzy C-means clustering; pixel space mapping; quick subtractive clustering; Clustering algorithms; Color; Computational complexity; Grid computing; Hypercubes; Image converters; Image segmentation; Pixel;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Innovative Computing Information and Control, 2008. ICICIC '08. 3rd International Conference on
Conference_Location :
Dalian, Liaoning
Print_ISBN :
978-0-7695-3161-8
Electronic_ISBN :
978-0-7695-3161-8
Type :
conf
DOI :
10.1109/ICICIC.2008.143
Filename :
4603389
Link To Document :
بازگشت