Title :
Unsupervised texture segmentation using grey levels rank-vectors
Author :
Asselin de Beauville, D.B.J.P. ; Mraghni, M.-C.
Author_Institution :
Ecole Nat. Superieure d´Ingenieurs de Genie Chimique, Toulouse
Abstract :
Presents an unsupervised texture segmentation algorithm. This algorithm uses the grey level rank-vectors (GLRV) to extract the texture features, and the unsupervised generalized minimum volume ellipsoid (GMVE) clustering algorithm to cluster the texture features into different classes without a priori information about the number of textures. The accuracy for real textured images is over 82%
Keywords :
feature extraction; image segmentation; image texture; GLRV; GMVE clustering algorithm; algorithm; grey level rank-vectors; real textured images; texture feature; unsupervised generalized minimum volume ellipsoid clustering algorithm; unsupervised texture segmentation; Bismuth; Classification algorithms; Clustering algorithms; Gaussian processes; Humans; Image processing; Machine vision; Robustness; Stochastic processes; Testing;
Conference_Titel :
Signal Processing, 1996., 3rd International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-2912-0
DOI :
10.1109/ICSIGP.1996.566526
