Face recognition using RBF neural networks and wavelet transform

Bicheng Li; Hujun Yin

Face recognition using RBF neural networks and wavelet transform

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

Abstract

Recently, wavelet transform and image fusion mechanism have been used in face recognition to improve the performance. In this paper, we propose a new face recognition method based on wavelet transform and radial basis function (RBF) fusion network. Firstly, an image is decomposed with wavelet transform (WT) to three levels. Secondly, the Fisherface method is applied to three low-frequency sub-images respectively. Then, the individual classifiers are fused using the RBF neural network. Experimental results show that the proposed method outperforms both individual classifiers and the direct Fisherface method. © Springer-Verlag Berlin Heidelberg 2005.

Original language	English
Title of host publication	Lecture Notes in Computer Science\|Lect. Notes Comput. Sci.
Editors	J. Wang, X. Liao, Z. Yi
Publisher	Springer Nature
Pages	105-111
Number of pages	6
Volume	3497
Publication status	Published - 2005
Event	Second International Symposium on Neural Networks: Advances in Neural Networks - ISNN 2005 - Chongqing Duration: 1 Jul 2005 → …

Conference

Conference	Second International Symposium on Neural Networks: Advances in Neural Networks - ISNN 2005
City	Chongqing
Period	1/07/05 → …

Cite this

@inproceedings{be70841b4df3465997cd2c638120ae82,

title = "Face recognition using RBF neural networks and wavelet transform",

abstract = "Recently, wavelet transform and image fusion mechanism have been used in face recognition to improve the performance. In this paper, we propose a new face recognition method based on wavelet transform and radial basis function (RBF) fusion network. Firstly, an image is decomposed with wavelet transform (WT) to three levels. Secondly, the Fisherface method is applied to three low-frequency sub-images respectively. Then, the individual classifiers are fused using the RBF neural network. Experimental results show that the proposed method outperforms both individual classifiers and the direct Fisherface method. {\textcopyright} Springer-Verlag Berlin Heidelberg 2005.",

author = "Bicheng Li and Hujun Yin",

year = "2005",

language = "English",

volume = "3497",

pages = "105--111",

editor = "J. Wang and X. Liao and Z. Yi",

booktitle = "Lecture Notes in Computer Science|Lect. Notes Comput. Sci.",

publisher = "Springer Nature",

address = "United States",

note = "Second International Symposium on Neural Networks: Advances in Neural Networks - ISNN 2005 ; Conference date: 01-07-2005",

}

TY - GEN

T1 - Face recognition using RBF neural networks and wavelet transform

AU - Li, Bicheng

AU - Yin, Hujun

PY - 2005

Y1 - 2005

N2 - Recently, wavelet transform and image fusion mechanism have been used in face recognition to improve the performance. In this paper, we propose a new face recognition method based on wavelet transform and radial basis function (RBF) fusion network. Firstly, an image is decomposed with wavelet transform (WT) to three levels. Secondly, the Fisherface method is applied to three low-frequency sub-images respectively. Then, the individual classifiers are fused using the RBF neural network. Experimental results show that the proposed method outperforms both individual classifiers and the direct Fisherface method. © Springer-Verlag Berlin Heidelberg 2005.

AB - Recently, wavelet transform and image fusion mechanism have been used in face recognition to improve the performance. In this paper, we propose a new face recognition method based on wavelet transform and radial basis function (RBF) fusion network. Firstly, an image is decomposed with wavelet transform (WT) to three levels. Secondly, the Fisherface method is applied to three low-frequency sub-images respectively. Then, the individual classifiers are fused using the RBF neural network. Experimental results show that the proposed method outperforms both individual classifiers and the direct Fisherface method. © Springer-Verlag Berlin Heidelberg 2005.

M3 - Conference contribution

VL - 3497

SP - 105

EP - 111

BT - Lecture Notes in Computer Science|Lect. Notes Comput. Sci.

A2 - Wang, J.

A2 - Liao, X.

A2 - Yi, Z.

PB - Springer Nature

T2 - Second International Symposium on Neural Networks: Advances in Neural Networks - ISNN 2005

Y2 - 1 July 2005

ER -

Face recognition using RBF neural networks and wavelet transform

Abstract

Conference

Fingerprint

Cite this