Skip to Main content Skip to Navigation

De l'appariement a l'indexation des images

Patrick Gros 1
1 MOVI - Modeling, localization, recognition and interpretation in computer vision
GRAVIR - IMAG - Laboratoire d'informatique GRAphique, VIsion et Robotique de Grenoble, Inria Grenoble - Rhône-Alpes, CNRS - Centre National de la Recherche Scientifique : FR71
Abstract : This document present all our work concerning image matching, modeling and indexing. This set of techniques was developed to build a system of automatic image recognition. In the first part, several matching techniques are presented. Each of these techniques is specifically devoted to a kind of images, respectively structured, gray level textured and color textured images. It is possible to make all these techniques collaborate to treat difficult images. The second part is devoted to the problem of modeling images and visual concepts. The basic tool is a hierarchical clustering algorithm, which uses an entropy based method to determine the number of cluster that should be formed. Image indexing is the subject of the last part. Two cases are studied: in the first one, the images are stored in the (\sc ram) memory of the computer, and several results of complexity are provided, and in the second case, the images are stored in the auxiliary memory of the computer. All the techniques presented are illustrated by examples using real images.
Document type :
Complete list of metadatas
Contributor : Thèses Imag <>
Submitted on : Thursday, February 19, 2004 - 2:42:11 PM
Last modification on : Monday, December 28, 2020 - 3:44:02 PM
Long-term archiving on: : Friday, April 2, 2010 - 8:33:39 PM


  • HAL Id : tel-00004889, version 1




Patrick Gros. De l'appariement a l'indexation des images. Interface homme-machine [cs.HC]. Institut National Polytechnique de Grenoble - INPG, 1998. Français. ⟨tel-00004889⟩



Record views


Files downloads