A Fast Algorithm for Image Deconvolution Based on a Rank Constrained Inverse Matrix Approximation Problem

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

3 Citas (Scopus)

Resumen

In this paper, we present a fast method for image deconvolution, which is based on the rank constrained inverse matrix approximation (RCIMA) problem. The RCIMA problem is a general case of the low-rank approximation problem proposed by Eckart-Young. This new algorithm, so-called the fast-RCIMA method, is based on tensor product and Tikhonov’s regularization to approximate the pseudoinverse and bilateral random projections to estimate the rank constrained approximation. The fast-RCIMA method reduces the execution time to estimate optimal solution and preserves the same accuracy of classical methods. We use training data as a substitute for knowledge of a forward model. Numerical simulations on measuring execution time and speedup confirmed the efficiency of the proposed method.

Idioma originalInglés
Título de la publicación alojadaProceedings of 6th International Congress on Information and Communication Technology, ICICT 2021
EditoresXin-She Yang, Simon Sherratt, Nilanjan Dey, Amit Joshi
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas165-176
Número de páginas12
ISBN (versión impresa)9789811623790
DOI
EstadoPublicada - 2022
Evento6th International Congress on Information and Communication Technology, ICICT 2021 - Virtual, Online
Duración: 25 feb 202126 feb 2021

Serie de la publicación

NombreLecture Notes in Networks and Systems
Volumen236
ISSN (versión impresa)2367-3370
ISSN (versión digital)2367-3389

Conferencia

Conferencia6th International Congress on Information and Communication Technology, ICICT 2021
CiudadVirtual, Online
Período25/02/2126/02/21

Huella

Profundice en los temas de investigación de 'A Fast Algorithm for Image Deconvolution Based on a Rank Constrained Inverse Matrix Approximation Problem'. En conjunto forman una huella única.

Citar esto