Iterative Poisson-Gaussian Noise Parametric Estimation for Blind Image Denoising

Abstract : This paper deals with noise parameter estimation from a single im- age under Poisson-Gaussian noise statistics. The problem is formu- lated within a mixed discrete-continuous optimization framework. The proposed approach jointly estimates the signal of interest and the noise parameters. This is achieved by introducing an adjustable reg- ularization term inside an optimized criterion, together with a data fidelity error measure. The optimal solution is sought iteratively by alternating the minimization of a label field and of a noise param- eter vector. Noise parameters are updated at each iteration using an Expectation-Maximization approach. The proposed algorithm is inspired from a spatial regularization approach for vector quantiza- tion. We illustrate the usefulness of our approach on macroconfocal images. The identified noise parameters are applied to a denoising algorithm, so yielding a fully automatic denoising scheme.
Type de document :
Communication dans un congrès
IEEE International Conference on Image Processing, Oct 2014, Paris, France. pp.1-5, 2014
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01060081
Contributeur : Anna Jezierska <>
Soumis le : jeudi 18 septembre 2014 - 09:57:01
Dernière modification le : jeudi 18 septembre 2014 - 10:21:55
Document(s) archivé(s) le : vendredi 19 décembre 2014 - 11:37:04

Fichier

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

Identifiants

  • HAL Id : hal-01060081, version 2

Citation

Anna Jezierska, Jean-Christophe Pesquet, Hugues Talbot, Caroline Chaux. Iterative Poisson-Gaussian Noise Parametric Estimation for Blind Image Denoising. IEEE International Conference on Image Processing, Oct 2014, Paris, France. pp.1-5, 2014. <hal-01060081v2>

Partager

Métriques

Consultations de
la notice

268

Téléchargements du document

182