DocumentCode :
432454
Title :
Color space selection for unsupervised color image segmentation by histogram multi-thresholding
Author :
Busin, L. ; Vandehbroucke, N. ; Macaire, L. ; Postaire, J.-G.
Author_Institution :
Departement Automatique, Ecole d´´Ingenieurs du Pas-de-Calais, Longuenesse, France
Volume :
1
fYear :
2004
fDate :
24-27 Oct. 2004
Firstpage :
203
Abstract :
We propose a new color image segmentation algorithm by unsupervised classification of pixels. This procedure iteratively constructs the classes by histogram multi-thresholding. For this purpose, the procedure selects different color spaces in which the modes of the 1D-histograms are separated as well as possible, so that each mode corresponds effectively to a region in the image.
Keywords :
image classification; image colour analysis; image segmentation; iterative methods; statistical analysis; unsupervised learning; 1D-histograms; color space selection; color spaces; histogram multithresholding; iterative construction; unsupervised color image segmentation; unsupervised pixel classification; Classification algorithms; Computer hacking; Histograms; Image analysis; Image color analysis; Image segmentation; Iterative algorithms; Pixel; Space technology;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing, 2004. ICIP '04. 2004 International Conference on
ISSN :
1522-4880
Print_ISBN :
0-7803-8554-3
Type :
conf
DOI :
10.1109/ICIP.2004.1418725
Filename :
1418725
Link To Document :
بازگشت