A Majorize-Minimize subspace approach for l2-l0 image regularization

Emilie Chouzenoux 1 Anna Jezierska 1 Jean-Christophe Pesquet 1 Hugues Talbot 1
1 LIGM - Laboratoire d'Informatique Gaspard-Monge
Abstract : In this work, we consider a class of differentiable criteria for sparse image computing problems, where a non-convex regularization is applied to an arbitrary linear transform of the target image. As special cases, it includes edge preserving measures or frame analysis potentials commonly used in image processing. As shown by our asymptotic results, the l2-l0 penalties we consider may be employed to provide approximate solutions to l0-penalized optimization problems. One of the advantages of the proposed approach is that it allows us to derive an efficient Majorize-Minimize subspace algorithm. The convergence of the algorithm is investigated by using recent results in non-convex optimization. The fast convergence properties of the proposed optimization method are illustrated through image processing examples. In particular, its effectiveness is demonstrated on several data recovery problems.
Type de document :
Article dans une revue
SIAM Journal on Imaging Sciences, Society for Industrial and Applied Mathematics, 2013, 6 (1), pp.563-591. 〈10.1137/11085997X〉
Liste complète des métadonnées
https://hal.archives-ouvertes.fr/hal-00789962
Contributeur : Emilie Chouzenoux <>
Soumis le : mardi 19 février 2013 - 10:31:30
Dernière modification le : lundi 3 décembre 2018 - 15:46:04

Lien texte intégral

arxiv

Identifiants

Collections

LIGM_SIGNAL | ENPC | LIGM | PARISTECH | LIGM_A3SI | UNIV-MLV | UPEC-UPEM | INSMI | TDS-MACS | ESIEE-PARIS

Citation

Emilie Chouzenoux, Anna Jezierska, Jean-Christophe Pesquet, Hugues Talbot. A Majorize-Minimize subspace approach for l2-l0 image regularization. SIAM Journal on Imaging Sciences, Society for Industrial and Applied Mathematics, 2013, 6 (1), pp.563-591. 〈10.1137/11085997X〉. 〈hal-00789962〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

591

Contact

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