A Robust approach based on fractional variable-order Total Variation and anisotropic diffusion for Poisson image restoration

Résumé

Abstract In this work, we propose a novel approach for Poisson image denoising based on a fractional variable-order total variation functional combined with an anisotropic diffusion operator. Poisson noise is particularly challenging to remove since it depends heavily on the image intensity. Due to the region geometry in images, efficient regularization terms capable of preserving piecewise smoothness in details and features such as Total Variation technique are necessary. The proposed model is discretized, and the Alternating Direction Method of Multipliers algorithm is used to split our optimization variables. We conduct various numerical experiments to demonstrate the superiority of our proposed approach over state-of-the-art methods in Poisson image denoising.

Citer ce document

Accès au document

Texte intégral en lecture en ligne, réservé aux abonnés SPHAERO et aux membres de l'institution. Se connecter

Voir l'article sur le site de la revue

Auteur(s)

Statistiques

Consultations : 1

Téléchargements : 0