Investigation on different pre-processing approaches for face recognition system

Maizura Mohd Sani, Khairul Anuar Ishak, Salina Abdul Samad

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

One of the challenges in face recognition system is to deal with inhomogeneous intensity problem that occur with different lighting conditions. In this paper, comparisons are made on several pre-processing methods i.e. histogram equalization, local binary pattern, wavelet transform and multiscale retinex. First, the input image is pre-processed with the illumination correction method before the classification task is done. The results are evaluated using the Yale, ORL and our own UKM database. These databases include images with various illumination conditions and expressions. Using PCA as the feature extraction and Euclidean Distance as the classification purposed, our experiments shows that multiscale retinex achieved the lowest equal error rates with 5.03% followed by local binary pattern (7.52%), wavelet transform (12.53%) and histogram equalization (12.97%) on average for all three databases.

Original languageEnglish
Title of host publication2nd International Conference on Computer Research and Development, ICCRD 2010
Pages692-695
Number of pages4
DOIs
Publication statusPublished - 2010
Event2nd International Conference on Computer Research and Development, ICCRD 2010 - Kuala Lumpur
Duration: 7 May 201010 May 2010

Other

Other2nd International Conference on Computer Research and Development, ICCRD 2010
CityKuala Lumpur
Period7/5/1010/5/10

Fingerprint

Face recognition
Lighting
Wavelet transforms
Processing
Feature extraction
Experiments

Keywords

  • Component
  • Illumination correction
  • Local binary pattern
  • Retinex
  • Wavelet transform

ASJC Scopus subject areas

  • Computer Science (miscellaneous)

Cite this

Mohd Sani, M., Ishak, K. A., & Abdul Samad, S. (2010). Investigation on different pre-processing approaches for face recognition system. In 2nd International Conference on Computer Research and Development, ICCRD 2010 (pp. 692-695). [5489531] https://doi.org/10.1109/ICCRD.2010.159

Investigation on different pre-processing approaches for face recognition system. / Mohd Sani, Maizura; Ishak, Khairul Anuar; Abdul Samad, Salina.

2nd International Conference on Computer Research and Development, ICCRD 2010. 2010. p. 692-695 5489531.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mohd Sani, M, Ishak, KA & Abdul Samad, S 2010, Investigation on different pre-processing approaches for face recognition system. in 2nd International Conference on Computer Research and Development, ICCRD 2010., 5489531, pp. 692-695, 2nd International Conference on Computer Research and Development, ICCRD 2010, Kuala Lumpur, 7/5/10. https://doi.org/10.1109/ICCRD.2010.159
Mohd Sani M, Ishak KA, Abdul Samad S. Investigation on different pre-processing approaches for face recognition system. In 2nd International Conference on Computer Research and Development, ICCRD 2010. 2010. p. 692-695. 5489531 https://doi.org/10.1109/ICCRD.2010.159
Mohd Sani, Maizura ; Ishak, Khairul Anuar ; Abdul Samad, Salina. / Investigation on different pre-processing approaches for face recognition system. 2nd International Conference on Computer Research and Development, ICCRD 2010. 2010. pp. 692-695
@inproceedings{51348deb32c3425d9cbd61d2db36d47b,
title = "Investigation on different pre-processing approaches for face recognition system",
abstract = "One of the challenges in face recognition system is to deal with inhomogeneous intensity problem that occur with different lighting conditions. In this paper, comparisons are made on several pre-processing methods i.e. histogram equalization, local binary pattern, wavelet transform and multiscale retinex. First, the input image is pre-processed with the illumination correction method before the classification task is done. The results are evaluated using the Yale, ORL and our own UKM database. These databases include images with various illumination conditions and expressions. Using PCA as the feature extraction and Euclidean Distance as the classification purposed, our experiments shows that multiscale retinex achieved the lowest equal error rates with 5.03{\%} followed by local binary pattern (7.52{\%}), wavelet transform (12.53{\%}) and histogram equalization (12.97{\%}) on average for all three databases.",
keywords = "Component, Illumination correction, Local binary pattern, Retinex, Wavelet transform",
author = "{Mohd Sani}, Maizura and Ishak, {Khairul Anuar} and {Abdul Samad}, Salina",
year = "2010",
doi = "10.1109/ICCRD.2010.159",
language = "English",
isbn = "9780769540436",
pages = "692--695",
booktitle = "2nd International Conference on Computer Research and Development, ICCRD 2010",

}

TY - GEN

T1 - Investigation on different pre-processing approaches for face recognition system

AU - Mohd Sani, Maizura

AU - Ishak, Khairul Anuar

AU - Abdul Samad, Salina

PY - 2010

Y1 - 2010

N2 - One of the challenges in face recognition system is to deal with inhomogeneous intensity problem that occur with different lighting conditions. In this paper, comparisons are made on several pre-processing methods i.e. histogram equalization, local binary pattern, wavelet transform and multiscale retinex. First, the input image is pre-processed with the illumination correction method before the classification task is done. The results are evaluated using the Yale, ORL and our own UKM database. These databases include images with various illumination conditions and expressions. Using PCA as the feature extraction and Euclidean Distance as the classification purposed, our experiments shows that multiscale retinex achieved the lowest equal error rates with 5.03% followed by local binary pattern (7.52%), wavelet transform (12.53%) and histogram equalization (12.97%) on average for all three databases.

AB - One of the challenges in face recognition system is to deal with inhomogeneous intensity problem that occur with different lighting conditions. In this paper, comparisons are made on several pre-processing methods i.e. histogram equalization, local binary pattern, wavelet transform and multiscale retinex. First, the input image is pre-processed with the illumination correction method before the classification task is done. The results are evaluated using the Yale, ORL and our own UKM database. These databases include images with various illumination conditions and expressions. Using PCA as the feature extraction and Euclidean Distance as the classification purposed, our experiments shows that multiscale retinex achieved the lowest equal error rates with 5.03% followed by local binary pattern (7.52%), wavelet transform (12.53%) and histogram equalization (12.97%) on average for all three databases.

KW - Component

KW - Illumination correction

KW - Local binary pattern

KW - Retinex

KW - Wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=77955200222&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955200222&partnerID=8YFLogxK

U2 - 10.1109/ICCRD.2010.159

DO - 10.1109/ICCRD.2010.159

M3 - Conference contribution

AN - SCOPUS:77955200222

SN - 9780769540436

SP - 692

EP - 695

BT - 2nd International Conference on Computer Research and Development, ICCRD 2010

ER -