On image restoration problems using new conjugate gradient methods

Autor:	Basim A. Hassan, Haneen A. Alashoor
Rok vydání:	2023
Předmět:	Control and Optimization Convergence property Image restoration problems Computer Networks and Communications Hardware and Architecture Descent property Signal Processing Conjugate gradient method Electrical and Electronic Engineering New coefficient conjugate Information Systems
Zdroj:	Indonesian Journal of Electrical Engineering and Computer Science. 29:1438
ISSN:	2502-4760 2502-4752
DOI:	10.11591/ijeecs.v29.i3.pp1438-1445
Popis:	The nonlinear conjugate gradient algorithm is one of the effective algorithms for optimization since it has low storage and simple structure properties. The coefficient conjugate is the basis of conjugate gradient algorithms with the desirable conjugate property. In this manuscript, we have derived a new second order information for the Hessian from objective function, which can give a new search direction. Based on new search direction, we have proposed the update formula interesting and nonlinear conjugate gradient method. Under wolfe line search and mild assumptions on objective function, the method possess sufficient descent property and are always globally convergent. Numerical results show that the method is effective and competitive to recover the original image from an image corrupted by impulse noise.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::67cfed114d69aa350836d4e7dba960fa https://doi.org/10.11591/ijeecs.v29.i3.pp1438-1445 Zobrazit plný text záznamu