Cover Image

Best Wavelet Function for Face Recognition Using Multi-Level Decomposition

The selection of appropriate wavelets is an important target for any application. In this paper, wavelets functions are examined in order to choose the best wavelet for face classification process and for finding the optimal number of levels of decomposition. Seven wavelet functions namely Symelt, D...

Full description

Saved in:
Bibliographic Details
Main Authors: Brahim Belhaouari Samir, BBS, Nadir Nourain, NN
Format: Article
Published: 2011
Subjects:
QA75 Electronic computers. Computer science
Online Access:http://eprints.utp.edu.my/7141/1/A7_PresentSchedule_071011.pdf
http://eprints.utp.edu.my/7141/
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.utp.eprints.7141
record_format eprints
spelling my.utp.eprints.71412017-01-19T08:22:22Z Best Wavelet Function for Face Recognition Using Multi-Level Decomposition Brahim Belhaouari Samir, BBS Nadir Nourain, NN QA75 Electronic computers. Computer science The selection of appropriate wavelets is an important target for any application. In this paper, wavelets functions are examined in order to choose the best wavelet for face classification process and for finding the optimal number of levels of decomposition. Seven wavelet functions namely Symelt, Daubechig, Coiflets, Mayer Discrete, Biorthogonal, Reverse Biorthogonal and Haar were tested with different number of decomposition levels and different number of biggest coefficients is selected to reduce the huge feature dimension, and then the Euclidean Distance Method (EDM) was used for classification process. Also a statistical method has been proposed to produce new metric of features coefficients, the experiments brought about 40% improvements in comparison to the method that accounts the biggest coefficients from the four levels of decompositions. The experiments have been performed on Olivetti Research Laboratory database (ORL) and Yale University database (YALE). The result showed the effect of wavelets proprieties on classification process and the Symelt wavelets are the optimum wavelets for the face classification with four levels. 2011-09-26 Article PeerReviewed application/pdf http://eprints.utp.edu.my/7141/1/A7_PresentSchedule_071011.pdf Brahim Belhaouari Samir, BBS and Nadir Nourain, NN (2011) Best Wavelet Function for Face Recognition Using Multi-Level Decomposition. IEEE International Conference on Research and Innovation Systems . http://eprints.utp.edu.my/7141/
institution Universiti Teknologi Petronas
building UTP Resource Centre
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Petronas
content_source UTP Institutional Repository
url_provider http://eprints.utp.edu.my/
topic QA75 Electronic computers. Computer science
spellingShingle QA75 Electronic computers. Computer science
Brahim Belhaouari Samir, BBS
Nadir Nourain, NN
Best Wavelet Function for Face Recognition Using Multi-Level Decomposition
description The selection of appropriate wavelets is an important target for any application. In this paper, wavelets functions are examined in order to choose the best wavelet for face classification process and for finding the optimal number of levels of decomposition. Seven wavelet functions namely Symelt, Daubechig, Coiflets, Mayer Discrete, Biorthogonal, Reverse Biorthogonal and Haar were tested with different number of decomposition levels and different number of biggest coefficients is selected to reduce the huge feature dimension, and then the Euclidean Distance Method (EDM) was used for classification process. Also a statistical method has been proposed to produce new metric of features coefficients, the experiments brought about 40% improvements in comparison to the method that accounts the biggest coefficients from the four levels of decompositions. The experiments have been performed on Olivetti Research Laboratory database (ORL) and Yale University database (YALE). The result showed the effect of wavelets proprieties on classification process and the Symelt wavelets are the optimum wavelets for the face classification with four levels.
format Article
author Brahim Belhaouari Samir, BBS
Nadir Nourain, NN
author_facet Brahim Belhaouari Samir, BBS
Nadir Nourain, NN
author_sort Brahim Belhaouari Samir, BBS
title Best Wavelet Function for Face Recognition Using Multi-Level Decomposition
title_short Best Wavelet Function for Face Recognition Using Multi-Level Decomposition
title_full Best Wavelet Function for Face Recognition Using Multi-Level Decomposition
title_fullStr Best Wavelet Function for Face Recognition Using Multi-Level Decomposition
title_full_unstemmed Best Wavelet Function for Face Recognition Using Multi-Level Decomposition
title_sort best wavelet function for face recognition using multi-level decomposition
publishDate 2011
url http://eprints.utp.edu.my/7141/1/A7_PresentSchedule_071011.pdf
http://eprints.utp.edu.my/7141/
_version_ 1738655549512220672
score 13.211869

Similar Items