A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization

S. Gendy; G. Kothapalli; A. Bouzerdoum

doi:10.1109/ANZIIS.2001.974060

A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization

S. Gendy, G. Kothapalli, A. Bouzerdoum

Edith Cowan University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

This paper presents a fast algorithm for a recurrent neural network that can restore a degraded image with fewer iterations and shorter processing time by using bound-constrained quadratic optimization (BCQO) and a weighted mask. The BCQO technique has already been used in signal restoration, however implementation of this method in image restoration requires considerable memory and it is computationally expensive. The proposed algorithm replaces the weight matrix of the network with a much smaller mask, thus reducing the processing time and requiring much less memory space. This algorithm produces better results than those obtained by Wiener filter, and achieves image restoration with less iterations compared to a modified Hopfield neural network.

Original language	English
Title of host publication	ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	111-115
Number of pages	5
ISBN (Electronic)	1740520610, 9781740520614
DOIs	https://doi.org/10.1109/ANZIIS.2001.974060
Publication status	Published - 2001
Externally published	Yes
Event	7th Australian and New Zealand Intelligent Information Systems Conference, ANZIIS 2001 - Perth, Australia Duration: 18 Nov 2001 → 21 Nov 2001

Publication series

Name	ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference

Conference

Conference	7th Australian and New Zealand Intelligent Information Systems Conference, ANZIIS 2001
Country/Territory	Australia
City	Perth
Period	18/11/01 → 21/11/01

Access to Document

10.1109/ANZIIS.2001.974060

Cite this

Gendy, S., Kothapalli, G., & Bouzerdoum, A. (2001). A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization. In ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference (pp. 111-115). Article 974060 (ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ANZIIS.2001.974060

Gendy, S. ; Kothapalli, G. ; Bouzerdoum, A. / A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization. ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference. Institute of Electrical and Electronics Engineers Inc., 2001. pp. 111-115 (ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference).

@inproceedings{4fef51c9fa1d4935b70f682d8e371eaa,

title = "A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization",

abstract = "This paper presents a fast algorithm for a recurrent neural network that can restore a degraded image with fewer iterations and shorter processing time by using bound-constrained quadratic optimization (BCQO) and a weighted mask. The BCQO technique has already been used in signal restoration, however implementation of this method in image restoration requires considerable memory and it is computationally expensive. The proposed algorithm replaces the weight matrix of the network with a much smaller mask, thus reducing the processing time and requiring much less memory space. This algorithm produces better results than those obtained by Wiener filter, and achieves image restoration with less iterations compared to a modified Hopfield neural network.",

author = "S. Gendy and G. Kothapalli and A. Bouzerdoum",

note = "Publisher Copyright: {\textcopyright} 2001 ARCME, Univ of WA.; 7th Australian and New Zealand Intelligent Information Systems Conference, ANZIIS 2001 ; Conference date: 18-11-2001 Through 21-11-2001",

year = "2001",

doi = "10.1109/ANZIIS.2001.974060",

language = "English",

series = "ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "111--115",

booktitle = "ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference",

address = "United States",

}

Gendy, S, Kothapalli, G & Bouzerdoum, A 2001, A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization. in ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference., 974060, ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference, Institute of Electrical and Electronics Engineers Inc., pp. 111-115, 7th Australian and New Zealand Intelligent Information Systems Conference, ANZIIS 2001, Perth, Australia, 18/11/01. https://doi.org/10.1109/ANZIIS.2001.974060

A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization. / Gendy, S.; Kothapalli, G.; Bouzerdoum, A.
ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference. Institute of Electrical and Electronics Engineers Inc., 2001. p. 111-115 974060 (ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization

AU - Gendy, S.

AU - Kothapalli, G.

AU - Bouzerdoum, A.

PY - 2001

Y1 - 2001

N2 - This paper presents a fast algorithm for a recurrent neural network that can restore a degraded image with fewer iterations and shorter processing time by using bound-constrained quadratic optimization (BCQO) and a weighted mask. The BCQO technique has already been used in signal restoration, however implementation of this method in image restoration requires considerable memory and it is computationally expensive. The proposed algorithm replaces the weight matrix of the network with a much smaller mask, thus reducing the processing time and requiring much less memory space. This algorithm produces better results than those obtained by Wiener filter, and achieves image restoration with less iterations compared to a modified Hopfield neural network.

AB - This paper presents a fast algorithm for a recurrent neural network that can restore a degraded image with fewer iterations and shorter processing time by using bound-constrained quadratic optimization (BCQO) and a weighted mask. The BCQO technique has already been used in signal restoration, however implementation of this method in image restoration requires considerable memory and it is computationally expensive. The proposed algorithm replaces the weight matrix of the network with a much smaller mask, thus reducing the processing time and requiring much less memory space. This algorithm produces better results than those obtained by Wiener filter, and achieves image restoration with less iterations compared to a modified Hopfield neural network.

UR - http://www.scopus.com/inward/record.url?scp=84953713300&partnerID=8YFLogxK

U2 - 10.1109/ANZIIS.2001.974060

DO - 10.1109/ANZIIS.2001.974060

M3 - Conference contribution

AN - SCOPUS:84953713300

T3 - ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference

SP - 111

EP - 115

BT - ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 7th Australian and New Zealand Intelligent Information Systems Conference, ANZIIS 2001

Y2 - 18 November 2001 through 21 November 2001

ER -

Gendy S, Kothapalli G, Bouzerdoum A. A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization. In ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference. Institute of Electrical and Electronics Engineers Inc. 2001. p. 111-115. 974060. (ANZIIS 2001 - Proceedings of the 7th Australian and New Zealand Intelligent Information Systems Conference). doi: 10.1109/ANZIIS.2001.974060

A fast algorithm for image restoration using a recurrent neural network with bound-constrained quadratic optimization

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this