The algorithms based on the Bregman iterative regularization are known for eciently solving convex constraint optimization problems. In this paper, we introduce a second order derivative scheme for the class of Bregman algorithms. Its properties of convergence and stability are investigated by means of numerical evidences. Moreover, we apply the proposed scheme to an isotropic Total Variation (TV) problem arising out of the Magnetic Resonance Image (MRI) denoising. Experimental results confirm that our algorithm has good performance in terms of denoising quality, eectiveness and robustness.
A second order derivative scheme based on bregman algorithm class / Campagna, Rosanna; Crisci, Serena; Cuomo, Salvatore; Galletti, Ardelio; Marcellino, Livia. - 1776:(2016), p. 040007. (Intervento presentato al convegno 2nd International Conference on Numerical Computations: Theory and Algorithms, NUMTA 2016 tenutosi a ita nel 2016) [10.1063/1.4965319].
A second order derivative scheme based on bregman algorithm class
CAMPAGNA, ROSANNA;CRISCI, SERENA;CUOMO, SALVATORE;GALLETTI, ARDELIO;MARCELLINO, LIVIA
2016
Abstract
The algorithms based on the Bregman iterative regularization are known for eciently solving convex constraint optimization problems. In this paper, we introduce a second order derivative scheme for the class of Bregman algorithms. Its properties of convergence and stability are investigated by means of numerical evidences. Moreover, we apply the proposed scheme to an isotropic Total Variation (TV) problem arising out of the Magnetic Resonance Image (MRI) denoising. Experimental results confirm that our algorithm has good performance in terms of denoising quality, eectiveness and robustness.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
