Proceedings of the 9th Joint International Conference on Information Sciences (JCIS-06)

Image Restoration Using RO Learning Approach

Authors
Chunshien Li1, Chan-Hung Yeh, Jye Lee
1Department of Computer Science, National Univ. of Tainan
Corresponding Author
Chunshien Li
Available Online October 2006.
DOI
10.2991/jcis.2006.22How to use a DOI?
Keywords
machine-learning, random-optimization, adaptive filter, block processing.
Abstract

In this paper, a machine-leaning-based adaptive approach is proposed to restore image from Gaussian corruption. The well-known Random-Optimization (RO) learning method is used for training of the adaptive filter. With the merit of model-free computation of RO, the derivative information is not required. Combined with block processing technique, the proposed adaptive filtering approach possesses fast convergence, moderate computation and simplicity. The proposed adaptive filter shows excellent filtering performance for image restoration.

Copyright
© 2006, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

Volume Title
Proceedings of the 9th Joint International Conference on Information Sciences (JCIS-06)
Series
Advances in Intelligent Systems Research
Publication Date
October 2006
ISBN
10.2991/jcis.2006.22
ISSN
1951-6851
DOI
10.2991/jcis.2006.22How to use a DOI?
Copyright
© 2006, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

TY  - CONF
AU  - Chunshien Li
AU  - Chan-Hung Yeh
AU  - Jye Lee
PY  - 2006/10
DA  - 2006/10
TI  - Image Restoration Using RO Learning Approach
BT  - Proceedings of the 9th Joint International Conference on Information Sciences (JCIS-06)
PB  - Atlantis Press
SP  - 91
EP  - 94
SN  - 1951-6851
UR  - https://doi.org/10.2991/jcis.2006.22
DO  - 10.2991/jcis.2006.22
ID  - Li2006/10
ER  -