A penalized weighted least squares approach for restoring data corrupted with signal-dependent noise

Audrey Repetti 1 Emilie Chouzenoux 1 Jean-Christophe Pesquet 1
1 LIGM - Laboratoire d'Informatique Gaspard-Monge
Abstract : This paper addresses the problem of recovering an image degraded by a linear operator and corrupted with an additive Gaussian noise with a signal-dependent variance. The considered observation model arises in several digital imaging devices. To solve this problem, a variational approach is adopted relying on a weighted least squares criterion which is penalized by a non-smooth function. In this context, the choice of an efficient optimization algorithm remains a challenging task. We propose here to extend a recent primal-dual proximal splitting approach by introducing a preconditioning strategy that is shown to significantly speed up the algorithm convergence. The good performance of the proposed method is illustrated through image restoration examples.
keyword : preconditioning convex optimization image restoration denoising deconvolution total variation regularization
Type de document :
Communication dans un congrès
20th European Signal Processing Conference (EUSIPCO 2012), Aug 2012, Bucarest, Romania. pp.1553-1557, 2012
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00733786
Contributeur : Emilie Chouzenoux <>
Soumis le : mercredi 3 juillet 2013 - 16:33:11
Dernière modification le : vendredi 12 juillet 2013 - 19:16:57
Document(s) archivé(s) le : vendredi 4 octobre 2013 - 02:20:09

Fichier

eusipco2012.pdf
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

  • HAL Id : hal-00733786, version 1

Collections

UNIV-MLV | LIGM_SIGNAL | ENPC-LIGM | LIGM | UPEC-UPEM | PARISTECH | ENPC

Citation

Audrey Repetti, Emilie Chouzenoux, Jean-Christophe Pesquet. A penalized weighted least squares approach for restoring data corrupted with signal-dependent noise. 20th European Signal Processing Conference (EUSIPCO 2012), Aug 2012, Bucarest, Romania. pp.1553-1557, 2012. <hal-00733786>

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de
la notice

226

Téléchargements du document

223

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD