Face recognition using integrated discrete cosine transform and kernel fisher discriminant analysis

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

Abstract

In face recognition applications, the dimension of the sample space is usually larger than the number of the samples in a training set. As a result, Fisher Linear Discriminant Analysis (FLD) based methods suffers due to singularity problem (of scatter matrix). This situation is often referred as "small sample size" (SSS) problem. Moreover, FLD is a linear algorithm by nature. Hence, it fails to extract important information from nonlinear and complex data such as face image. To remedy this problem, this paper presents a new face recognition approach by integrating Discrete Cosine Transform (DCT) and Kernel Fisher Discriminant Analysis (KFDA). The DCT has the capability to compact the energy in an image and let the dimensionality of the input sample space to be reduced. Then, KFDA, a new variant of FLD, will be used to extract the most discriminating feature. This is performed by transforming the reduced DCT subset using a nonlinear kernel function to a high dimensional nonlinear feature space and then followed by the FLD step. Based on the extensive experiments performed on ORL Database, the highest recognition accuracy of 95.375% is achieved with only 24 features.

Original languageEnglish
Title of host publication2006 International Conference on Computing and Informatics, ICOCI '06
DOIs
Publication statusPublished - 01 Dec 2006
Event2006 International Conference on Computing and Informatics, ICOCI '06 - Kuala Lumpur, Malaysia
Duration: 06 Jun 200608 Jun 2006

Other

Other2006 International Conference on Computing and Informatics, ICOCI '06
CountryMalaysia
CityKuala Lumpur
Period06/06/0608/06/06

Fingerprint

Discrete cosine transforms
Discriminant analysis
Face recognition
Set theory
Experiments

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Software

Cite this

V. Janahiraman, Tiagrajah ; Omar, Jamaludin ; Farukh, H. N. / Face recognition using integrated discrete cosine transform and kernel fisher discriminant analysis. 2006 International Conference on Computing and Informatics, ICOCI '06. 2006.
@inproceedings{54b019e671c44e48ab0d7d71e5edd4f8,
title = "Face recognition using integrated discrete cosine transform and kernel fisher discriminant analysis",
abstract = "In face recognition applications, the dimension of the sample space is usually larger than the number of the samples in a training set. As a result, Fisher Linear Discriminant Analysis (FLD) based methods suffers due to singularity problem (of scatter matrix). This situation is often referred as {"}small sample size{"} (SSS) problem. Moreover, FLD is a linear algorithm by nature. Hence, it fails to extract important information from nonlinear and complex data such as face image. To remedy this problem, this paper presents a new face recognition approach by integrating Discrete Cosine Transform (DCT) and Kernel Fisher Discriminant Analysis (KFDA). The DCT has the capability to compact the energy in an image and let the dimensionality of the input sample space to be reduced. Then, KFDA, a new variant of FLD, will be used to extract the most discriminating feature. This is performed by transforming the reduced DCT subset using a nonlinear kernel function to a high dimensional nonlinear feature space and then followed by the FLD step. Based on the extensive experiments performed on ORL Database, the highest recognition accuracy of 95.375{\%} is achieved with only 24 features.",
author = "{V. Janahiraman}, Tiagrajah and Jamaludin Omar and Farukh, {H. N.}",
year = "2006",
month = "12",
day = "1",
doi = "10.1109/ICOCI.2006.5276535",
language = "English",
isbn = "1424402204",
booktitle = "2006 International Conference on Computing and Informatics, ICOCI '06",

}

V. Janahiraman, T, Omar, J & Farukh, HN 2006, Face recognition using integrated discrete cosine transform and kernel fisher discriminant analysis. in 2006 International Conference on Computing and Informatics, ICOCI '06., 5276535, 2006 International Conference on Computing and Informatics, ICOCI '06, Kuala Lumpur, Malaysia, 06/06/06. https://doi.org/10.1109/ICOCI.2006.5276535

Face recognition using integrated discrete cosine transform and kernel fisher discriminant analysis. / V. Janahiraman, Tiagrajah; Omar, Jamaludin; Farukh, H. N.

2006 International Conference on Computing and Informatics, ICOCI '06. 2006. 5276535.

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

TY - GEN

T1 - Face recognition using integrated discrete cosine transform and kernel fisher discriminant analysis

AU - V. Janahiraman, Tiagrajah

AU - Omar, Jamaludin

AU - Farukh, H. N.

PY - 2006/12/1

Y1 - 2006/12/1

N2 - In face recognition applications, the dimension of the sample space is usually larger than the number of the samples in a training set. As a result, Fisher Linear Discriminant Analysis (FLD) based methods suffers due to singularity problem (of scatter matrix). This situation is often referred as "small sample size" (SSS) problem. Moreover, FLD is a linear algorithm by nature. Hence, it fails to extract important information from nonlinear and complex data such as face image. To remedy this problem, this paper presents a new face recognition approach by integrating Discrete Cosine Transform (DCT) and Kernel Fisher Discriminant Analysis (KFDA). The DCT has the capability to compact the energy in an image and let the dimensionality of the input sample space to be reduced. Then, KFDA, a new variant of FLD, will be used to extract the most discriminating feature. This is performed by transforming the reduced DCT subset using a nonlinear kernel function to a high dimensional nonlinear feature space and then followed by the FLD step. Based on the extensive experiments performed on ORL Database, the highest recognition accuracy of 95.375% is achieved with only 24 features.

AB - In face recognition applications, the dimension of the sample space is usually larger than the number of the samples in a training set. As a result, Fisher Linear Discriminant Analysis (FLD) based methods suffers due to singularity problem (of scatter matrix). This situation is often referred as "small sample size" (SSS) problem. Moreover, FLD is a linear algorithm by nature. Hence, it fails to extract important information from nonlinear and complex data such as face image. To remedy this problem, this paper presents a new face recognition approach by integrating Discrete Cosine Transform (DCT) and Kernel Fisher Discriminant Analysis (KFDA). The DCT has the capability to compact the energy in an image and let the dimensionality of the input sample space to be reduced. Then, KFDA, a new variant of FLD, will be used to extract the most discriminating feature. This is performed by transforming the reduced DCT subset using a nonlinear kernel function to a high dimensional nonlinear feature space and then followed by the FLD step. Based on the extensive experiments performed on ORL Database, the highest recognition accuracy of 95.375% is achieved with only 24 features.

UR - http://www.scopus.com/inward/record.url?scp=71249128487&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=71249128487&partnerID=8YFLogxK

U2 - 10.1109/ICOCI.2006.5276535

DO - 10.1109/ICOCI.2006.5276535

M3 - Conference contribution

SN - 1424402204

SN - 9781424402205

BT - 2006 International Conference on Computing and Informatics, ICOCI '06

ER -