A Concentration-Based Adaptive Approach to Region Merging of Optimal Time and Space Complexities

Christophe Fiorio 1 Andre Mas 2
1 ICAR - Image & Interaction
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : In this paper, we investigate image segmentation as a statistical and computa- tional problem. The observed image is sampled from a theoretical, unknown image, in which pixels are represented by distributions. Our objective is to approximate as best as possible the region segmentation in the ideal image, where each region has pixels with identical expectations, but adjacent regions have different pixel’s expectations. From that model, a concentration-based statistical test for deciding region merging is built, limiting the risk of wrong merges. The analysis is carried out without any assumption on the distribu- tions: we avoid in particular the classics of variance analysis, normality and homocedasticity. A practical approximation of the test is given, of constant time and space computation, which leads in turn to a segmentation algorithm of optimal complexity, easy to implement. Some experiments on various types of images shed light on the quality of the segmentations obtained.
Type de document :
Communication dans un congrès
Proceedings of the British Machine Vision Conference, Sep 2000, Bristol, United Kingdom. BMVA Press, 2000, 〈10.5244/C.14.78〉
Liste complète des métadonnées

Littérature citée [6 références]  Voir  Masquer  Télécharger

https://hal-lirmm.ccsd.cnrs.fr/lirmm-01168062
Contributeur : Christophe Fiorio <>
Soumis le : jeudi 25 juin 2015 - 11:17:34
Dernière modification le : vendredi 22 juin 2018 - 14:00:02
Document(s) archivé(s) le : vendredi 9 octobre 2015 - 17:25:35

Fichier

FioNocBMVC00_A Concentration-B...
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Christophe Fiorio, Andre Mas. A Concentration-Based Adaptive Approach to Region Merging of Optimal Time and Space Complexities . Proceedings of the British Machine Vision Conference, Sep 2000, Bristol, United Kingdom. BMVA Press, 2000, 〈10.5244/C.14.78〉. 〈lirmm-01168062〉

Partager

Métriques

Consultations de la notice

85

Téléchargements de fichiers

200