A new multi-scale approach to the segmentation of multi-textured images
Eric Favier,
Laboratoire T. S. I. Equipe lngénierie de la Vision C.N.R.S. URA 842, 23 Rue du Docteur Paul Michelon 42023 Saint-Etienne CEDEX 2, France
Bernard Laget,
Laboratoire T. S. I. Equipe lngénierie de la Vision C.N.R.S. URA 842, 23 Rue du Docteur Paul Michelon 42023 Saint-Etienne CEDEX 2, France
Abstract
This article presents a new method for the segmentation of multi-textured images. This method, based on statistical criteria, is multi-scale. In fact the images are studying at any study scales and the best are automatically selected so that the segmentation can be done. The obtained results are in accordance with the segmentation realized by the human preattentive vision.
Keywords : multi-scale, segmentation, study scale, textured images
Pour citer cet article
Eric Favier & Bernard Laget, «A new multi-scale approach to the segmentation of multi-textured images», Acta Stereologica [En ligne], Volume 14 (1995), Number 2 - Quantitative image analysis - Nov. 1995, 179-184 URL : https://popups.uliege.be/0351-580x/index.php?id=832.