Image restoration using a kNN-variant of the mean-shift

Abstract : The image restoration problem is addressed in the variational framework. The focus was set on denoising. The statistics of natural images are consistent with the Markov random field principles. Therefore, a restoration process should preserve the correlation between adjacent pixels. The proposed approach minimizes the conditional entropy of a pixel knowing its neighborhood. The conditional aspect helps preserving local image structures such as edges and textures. The statistical properties of the degraded image are estimated using a novel, adaptive weighted k-th nearest neighbor (kNN) strategy. The derived gradient descent procedure is mainly based on meanshift computations in this framework.
Type de document :
Communication dans un congrès
IEEE International Conference on Image Processing, Oct 2008, San Diego, California, United States. 2008
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-00379329
Contributeur : Cesario Vincenzo Angelino <>
Soumis le : mercredi 27 octobre 2010 - 10:41:36
Dernière modification le : vendredi 29 octobre 2010 - 11:36:44
Document(s) archivé(s) le : jeudi 1 décembre 2016 - 21:25:36

Fichier

ICIP08mod.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00379329, version 2

Collections

Citation

Cesario Vincenzo Angelino, Eric Debreuve, Michel Barlaud. Image restoration using a kNN-variant of the mean-shift. IEEE International Conference on Image Processing, Oct 2008, San Diego, California, United States. 2008. <hal-00379329v2>

Partager

Métriques

Consultations de
la notice

149

Téléchargements du document

161