Skip to Main content Skip to Navigation
Theses

Diagrammes de Voronoi 2D et 3D, applications en analyse d'images

Etienne Bertin 1
1 INFODIS
TIMC - Techniques de l'Ingénierie Médicale et de la Complexité - Informatique, Mathématiques et Applications, Grenoble - UMR 5525
Abstract : The aim of this Ph.D. thesis is to show how computational geometry in the general case and the Voronoi diagram in the particular case contribute to image analysis. The purpose of image analysis is the description of the image content in order to interpret it and make a decision. Computational geometry is used to find optimal algorithms to solve geometric problems. We are interested here in the representation image problem. We will propose a solution based on the partitionning of the image content. Among the various solutions to partitionning, we will specifically develop the Voronoi model. The 3D shape codage by skeletons will then be developed as it relates to the generalized 3D Voronoi diagram. Image segmentation is an important problem in image analysis. We will present a new approach in a pyramidal environment, initialized by the Voronoi diagram and controled by Hopfiel neural networks. Finally, Markov random field will be used to determine an optimal partitionning in such a way which is still to be defined.
Complete list of metadata

Cited literature [140 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00005078
Contributor : Thèses Imag <>
Submitted on : Wednesday, February 25, 2004 - 10:57:41 AM
Last modification on : Thursday, November 19, 2020 - 3:56:09 PM
Long-term archiving on: : Friday, September 14, 2012 - 11:40:37 AM

Identifiers

  • HAL Id : tel-00005078, version 1

Collections

Citation

Etienne Bertin. Diagrammes de Voronoi 2D et 3D, applications en analyse d'images. Interface homme-machine [cs.HC]. Université Joseph-Fourier - Grenoble I, 1994. Français. ⟨tel-00005078⟩

Share

Metrics

Record views

1226

Files downloads

3051