A hybrid interior point - deep learning approach for Poisson image deblurring

Mathilde Galinier (1) , Marco Prato (1) , Emilie Chouzenoux (2) , Jean-Christophe Pesquet (2, 3)

1 UNIMORE - Università degli Studi di Modena e Reggio Emilia = University of Modena and Reggio Emilia
2 OPIS - OPtimisation Imagerie et Santé
3 CVN - Centre de vision numérique

In this paper we address the problem of deconvolution of an image corrupted with Poisson noise by reformulating the restoration process as a constrained minimization of a suitable regularized data fidelity function. The minimization step is performed by means of an interior-point approach, in which the constraints are incorporated within the objective function through a barrier penalty and a forward-backward algorithm is exploited to build a minimizing sequence. The key point of our proposed scheme is that the choice of the regularization, barrier and step-size parameters defining the interior point approach is automatically performed by a deep learning strategy. Numerical tests on Poisson corrupted benchmark datasets show that our method can obtain very good performance when compared to a state-of-the-art variational deblurring strategy.

Mots clés: neural network; deep unfolding; proximal algorithms; Interior point method; Poisson image restoration

Domaines: Traitement du signal et de l'image [eess.SP]

Mathilde Galinier, Marco Prato, Emilie Chouzenoux, Jean-Christophe Pesquet. A hybrid interior point - deep learning approach for Poisson image deblurring. MLSP 2020 - IEEE International Workshop on Machine Learning for Signal Processing, Sep 2020, Espoo, Finland. 6p. ⟨hal-03647254⟩