Débruitage de séquences vidéo en présence de perturbations fortement impulsives

Mohammed El Hassouni 1
1 LRIT
GSCM-LRIT - Laboratoire de Recherche en Informatique et Télécommunications [Rabat]
Abstract : In this document, we are interested to the video denoising in the presence of heavily impulsive perturbations. These additive perturbations which can occur during acquisition, transmission or compression of video ows cannot be modeled in an adequate way by a Gaussian distribution. To address this problem two types of solutions are generally adopted : parametric methods and non-parametric methods. In a first part, we propose to use the higher order statistics (HOS). The HOS-based algorithms are compared with the techniques based on the second order statistics (SOS). The experimental evaluation of the performances emphasizes the interest of such approach. In a second part, we propose to model the perturbation process by the α-stable distribution. The treatments resulting from this modeling show the eectiveness of the approach suggested in term of SNR and computational time gain.
Document type :
Theses
Complete list of metadatas

Cited literature [186 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/tel-01374673
Contributor : Mohammed El Hassouni <>
Submitted on : Friday, September 30, 2016 - 7:37:56 PM
Last modification on : Monday, October 3, 2016 - 1:48:38 PM
Long-term archiving on : Saturday, December 31, 2016 - 4:31:09 PM

Identifiers

  • HAL Id : tel-01374673, version 1

Citation

Mohammed El Hassouni. Débruitage de séquences vidéo en présence de perturbations fortement impulsives. Informatique [cs]. Université de Bourgogne, 2005. Français. ⟨NNT : 2005DIJOS004⟩. ⟨tel-01374673⟩

Share

Metrics

Record views

88

Files downloads

423