Iterative Image Restoration and the Stopping Criteria

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989592%3A15310%2F07%3A00003608" target="_blank" >RIV/61989592:15310/07:00003608 - isvavai.cz</a>

Result on the web

—

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Iterative Image Restoration and the Stopping Criteria

Original language description

Digital images can be negatively influenced by several aspects, mostly by camera movement, unfocussed lenses, and noise from the camera sensor. The reconstruction can be done by several reconstruction methods. The methods can be non-iterative such as Wiener reconstruction. The Wiener reconstruction is effective and fast for the most of ordinary images. Unfortunately, some artifacts are visible after the reconstruction. From the iterative methods, the Lucy-Richardson (LR) or blind deconvolution are sometimes used. Iterative algorithms are slower but the results can be usually better. The problem with iterative methods is that the number of iterations needed in order to achieve the image of adequate quality is not known ahead. Our experiments show how tolook for appropriate stopping criteria, and that in case of the LR deconvolution it is possible to find such criteria. The original non-degraded image was not available. The considerations were based on judging of 100 images. The experim

Czech name

Iterative Image Restoration and the Stopping Criteria

Czech description

Digital images can be negatively influenced by several aspects, mostly by camera movement, unfocussed lenses, and noise from the camera sensor. The reconstruction can be done by several reconstruction methods. The methods can be non-iterative such as Wiener reconstruction. The Wiener reconstruction is effective and fast for the most of ordinary images. Unfortunately, some artifacts are visible after the reconstruction. From the iterative methods, the Lucy-Richardson (LR) or blind deconvolution are sometimes used. Iterative algorithms are slower but the results can be usually better. The problem with iterative methods is that the number of iterations needed in order to achieve the image of adequate quality is not known ahead. Our experiments show how tolook for appropriate stopping criteria, and that in case of the LR deconvolution it is possible to find such criteria. The original non-degraded image was not available. The considerations were based on judging of 100 images. The experim

Type

D - Article in proceedings

CEP classification

JC - Computer hardware and software

OECD FORD branch

—

Project

Result was created during the realization of more than one project. More information in the Projects tab.

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2007

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Proceedings of the 11th World Multi-Conference on Systemics, Cybernetics and Informatics

ISBN

1-934272-16-7

ISSN

—

e-ISSN

—

Number of pages

370

Pages from-to

192-196

Publisher name

International Institute of Informatics and Systemics

Place of publication

Orlando, Florida, USA

Event location

—

Event date

—

Type of event by nationality

—

UT code for WoS article

—