Agenda

Fabio EL SAMRANI soutiendra sa thèse de doctorat intitulée : « Adaptive blind restoration for monochrome and hyperspectral images », sous la direction de Kacem Chedhi (PR, IETR - équipe Multip Lannion / Université de Rennes, directeur de travaux).

Venir à l'Enssat

Abstract:
Image restoration presents a significant challenge when regularization parameter values, PSF, and other a priori knowledge are not available. The objective of this thesis is to develop an easily applicable restoration method by eliminating the necessity for prior information and empirical parameter tuning. To achieve this objective, an adaptive blind image restoration method that operates without requiring prior information is therefore developed. This method can be applied to restore monochrome, multispectral, and hyperspectral images, while optimizing restoration results without the need for empirical parameter tuning.
The superiority of our proposed adaptive blind restoration method is demonstrated through evaluations on diverse image databases, outperforming eleven existing non-neural network and supervised/semi-supervised neural network methods from the state-of-the-art.
In conclusion, the proposed method can be easily applied to restore degraded images due to its blind nature.