Modular Dynamic RBF Neural Network for Face Recognition

Sue Inn Ch'ng, Kah Phooi Seng, Li-Minn Ang

Research output: Book chapter/Published conference paperConference paper

4 Citations (Scopus)

Abstract

Over the years, we have seen an increase in the use of RBF neural networks for the task of face recognition. However, the use of second order algorithms as the learning algorithm for all the adjustable parameters in such networks are rare due to the high computational complexity of the calculation of the Jacobian and Hessian matrix. Hence, in this paper, we propose a modular structural training architecture to adapt the Levenberg-Marquardt based RBF neural network for the application of face recognition. In addition to the proposal of the modular structural training architecture, we have also investigated the use of different front-end processors to reduce the dimension size of the feature vectors prior to its application to the LM-based RBF neural network. The investigative study was done on three standard face databases; ORL, Yale and AR databases.
Original languageEnglish
Title of host publicationProceedings of the 2012 IEEE Conference on Open Systems
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages133-138
Number of pages6
ISBN (Print)9781467310444
DOIs
Publication statusPublished - 2012
Event2012 IEEE Conference on Open Systems: ICOS 2012 - Grand Seasons Hotel, Kuala Lumpur, Malaysia
Duration: 21 Oct 201224 Oct 2012
https://web.archive.org/web/20120726011442/http://www.cs.ieeemalaysia.org/icos2012

Publication series

NameIEEE Conference on Open Systems

Conference

Conference2012 IEEE Conference on Open Systems
CountryMalaysia
CityKuala Lumpur
Period21/10/1224/10/12
Internet address

Cite this

Ch'ng, S. I., Seng, K. P., & Ang, L-M. (2012). Modular Dynamic RBF Neural Network for Face Recognition. In Proceedings of the 2012 IEEE Conference on Open Systems (pp. 133-138). [6417629] (IEEE Conference on Open Systems). United States: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICOS.2012.6417629
Ch'ng, Sue Inn ; Seng, Kah Phooi ; Ang, Li-Minn. / Modular Dynamic RBF Neural Network for Face Recognition. Proceedings of the 2012 IEEE Conference on Open Systems. United States : IEEE, Institute of Electrical and Electronics Engineers, 2012. pp. 133-138 (IEEE Conference on Open Systems).
@inproceedings{6a48639c93624a659b481d07bf31d39f,
title = "Modular Dynamic RBF Neural Network for Face Recognition",
abstract = "Over the years, we have seen an increase in the use of RBF neural networks for the task of face recognition. However, the use of second order algorithms as the learning algorithm for all the adjustable parameters in such networks are rare due to the high computational complexity of the calculation of the Jacobian and Hessian matrix. Hence, in this paper, we propose a modular structural training architecture to adapt the Levenberg-Marquardt based RBF neural network for the application of face recognition. In addition to the proposal of the modular structural training architecture, we have also investigated the use of different front-end processors to reduce the dimension size of the feature vectors prior to its application to the LM-based RBF neural network. The investigative study was done on three standard face databases; ORL, Yale and AR databases.",
keywords = "modular structure, RBF neural networks, Levenberg-Marquardt algorithm, face recognition",
author = "Ch'ng, {Sue Inn} and Seng, {Kah Phooi} and Li-Minn Ang",
year = "2012",
doi = "10.1109/ICOS.2012.6417629",
language = "English",
isbn = "9781467310444",
series = "IEEE Conference on Open Systems",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
pages = "133--138",
booktitle = "Proceedings of the 2012 IEEE Conference on Open Systems",
address = "United States",

}

Ch'ng, SI, Seng, KP & Ang, L-M 2012, Modular Dynamic RBF Neural Network for Face Recognition. in Proceedings of the 2012 IEEE Conference on Open Systems., 6417629, IEEE Conference on Open Systems, IEEE, Institute of Electrical and Electronics Engineers, United States, pp. 133-138, 2012 IEEE Conference on Open Systems, Kuala Lumpur, Malaysia, 21/10/12. https://doi.org/10.1109/ICOS.2012.6417629

Modular Dynamic RBF Neural Network for Face Recognition. / Ch'ng, Sue Inn; Seng, Kah Phooi; Ang, Li-Minn.

Proceedings of the 2012 IEEE Conference on Open Systems. United States : IEEE, Institute of Electrical and Electronics Engineers, 2012. p. 133-138 6417629 (IEEE Conference on Open Systems).

Research output: Book chapter/Published conference paperConference paper

TY - GEN

T1 - Modular Dynamic RBF Neural Network for Face Recognition

AU - Ch'ng, Sue Inn

AU - Seng, Kah Phooi

AU - Ang, Li-Minn

PY - 2012

Y1 - 2012

N2 - Over the years, we have seen an increase in the use of RBF neural networks for the task of face recognition. However, the use of second order algorithms as the learning algorithm for all the adjustable parameters in such networks are rare due to the high computational complexity of the calculation of the Jacobian and Hessian matrix. Hence, in this paper, we propose a modular structural training architecture to adapt the Levenberg-Marquardt based RBF neural network for the application of face recognition. In addition to the proposal of the modular structural training architecture, we have also investigated the use of different front-end processors to reduce the dimension size of the feature vectors prior to its application to the LM-based RBF neural network. The investigative study was done on three standard face databases; ORL, Yale and AR databases.

AB - Over the years, we have seen an increase in the use of RBF neural networks for the task of face recognition. However, the use of second order algorithms as the learning algorithm for all the adjustable parameters in such networks are rare due to the high computational complexity of the calculation of the Jacobian and Hessian matrix. Hence, in this paper, we propose a modular structural training architecture to adapt the Levenberg-Marquardt based RBF neural network for the application of face recognition. In addition to the proposal of the modular structural training architecture, we have also investigated the use of different front-end processors to reduce the dimension size of the feature vectors prior to its application to the LM-based RBF neural network. The investigative study was done on three standard face databases; ORL, Yale and AR databases.

KW - modular structure

KW - RBF neural networks

KW - Levenberg-Marquardt algorithm

KW - face recognition

U2 - 10.1109/ICOS.2012.6417629

DO - 10.1109/ICOS.2012.6417629

M3 - Conference paper

SN - 9781467310444

T3 - IEEE Conference on Open Systems

SP - 133

EP - 138

BT - Proceedings of the 2012 IEEE Conference on Open Systems

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - United States

ER -

Ch'ng SI, Seng KP, Ang L-M. Modular Dynamic RBF Neural Network for Face Recognition. In Proceedings of the 2012 IEEE Conference on Open Systems. United States: IEEE, Institute of Electrical and Electronics Engineers. 2012. p. 133-138. 6417629. (IEEE Conference on Open Systems). https://doi.org/10.1109/ICOS.2012.6417629