Title :
Segmentation of color image by ϕβ criterion fuzzy theory
Author :
El Matouat, Abdelaziz ; Hamzaoui, Hassania ; Martin, Patrick
Author_Institution :
CERENE, Le Havre Univ.
Abstract :
In this paper, we propose to use the information criterion ϕβ to identify the optimal number of clusters in the segmentation of a color image. The performance of this criterion is verified on the image test "House", "Monarch", "Lenna" and "Peppers", and compared with the selection obtained by the Chen and Lu fuzzy segmentation method. We verify that the new proposed method is efficient with respect to Chen\´s algorithm. We finally propose an appropriate choice for the radius in order to have an optimal segmentation of the image
Keywords :
fuzzy set theory; image colour analysis; image segmentation; ϕβ criterion; color image segmentation; fuzzy theory; Clustering algorithms; Colored noise; Fuzzy sets; Histograms; Image color analysis; Image segmentation; Maximum likelihood detection; Maximum likelihood estimation; Pattern recognition; Pixel; ϕβ criterion; Fuzzy clustering algorithm; Fuzzy sets; Histogram; Information criteria;
Conference_Titel :
IEEE Industrial Electronics, IECON 2006 - 32nd Annual Conference on
Conference_Location :
Paris
Print_ISBN :
1-4244-0390-1
DOI :
10.1109/IECON.2006.347677
