Face biometrics based on principal component analysis and linear discriminant analysis
Problem statement: In facial biometrics, face features are used as the required human traits for automatic recognition. Feature extracted from face images are significant for face biometrics system performance. Approach: In this thesis, a framework of facial biometric was designed based on two subsp...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Science Publications
2010
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/26149/2/jcssp.2010.693.699 |
| _version_ | 1796856150830874624 |
|---|---|
| author | Shaikh Salleh, Sheikh Hussain L. H., Chan C. M., Ting |
| author_facet | Shaikh Salleh, Sheikh Hussain L. H., Chan C. M., Ting |
| author_sort | Shaikh Salleh, Sheikh Hussain |
| collection | ePrints |
| description | Problem statement: In facial biometrics, face features are used as the required human traits for automatic recognition. Feature extracted from face images are significant for face biometrics system performance. Approach: In this thesis, a framework of facial biometric was designed based on two subspace methods i.e., Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). First, PCA is used for dimension reduction, where original face images are projected into lower-dimensional face representations. Second, LDA was proposed to provide a solution of better discriminant. Both PCA and LDA features were presented to Euclidean distance measurement which is conveniently used as a benchmark. The algorithms were evaluated in face identification and verification using a standard face database-AT and T and a locally collected database-CBE. Each database consists of 400 images and 320 images respectively. Results: LDA-based methods outperform PCA for both face identification and verification. For face identification, PCA achieves accuracy of 91.9% (AT and T) and 76.7% (CBE) while LDA 94.2% (AT and T) and 83.1% (CBE). For face verification, PCA achieves Equal Error Rate (EER) of 1.15% (AT and T), 7.3% (CBE) while LDA 0.78% (AT and T) and 5.81% (CBE). Conclusion/Recommendations: This study had proved that, when given sufficient training samples, LDA is able to provide better discriminant ability in feature extraction for face biometrics. |
| first_indexed | 2024-03-05T18:39:12Z |
| format | Article |
| id | utm.eprints-26149 |
| institution | Universiti Teknologi Malaysia - ePrints |
| language | English |
| last_indexed | 2024-03-05T18:39:12Z |
| publishDate | 2010 |
| publisher | Science Publications |
| record_format | dspace |
| spelling | utm.eprints-261492017-10-13T12:30:26Z http://eprints.utm.my/26149/ Face biometrics based on principal component analysis and linear discriminant analysis Shaikh Salleh, Sheikh Hussain L. H., Chan C. M., Ting QA75 Electronic computers. Computer science Problem statement: In facial biometrics, face features are used as the required human traits for automatic recognition. Feature extracted from face images are significant for face biometrics system performance. Approach: In this thesis, a framework of facial biometric was designed based on two subspace methods i.e., Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). First, PCA is used for dimension reduction, where original face images are projected into lower-dimensional face representations. Second, LDA was proposed to provide a solution of better discriminant. Both PCA and LDA features were presented to Euclidean distance measurement which is conveniently used as a benchmark. The algorithms were evaluated in face identification and verification using a standard face database-AT and T and a locally collected database-CBE. Each database consists of 400 images and 320 images respectively. Results: LDA-based methods outperform PCA for both face identification and verification. For face identification, PCA achieves accuracy of 91.9% (AT and T) and 76.7% (CBE) while LDA 94.2% (AT and T) and 83.1% (CBE). For face verification, PCA achieves Equal Error Rate (EER) of 1.15% (AT and T), 7.3% (CBE) while LDA 0.78% (AT and T) and 5.81% (CBE). Conclusion/Recommendations: This study had proved that, when given sufficient training samples, LDA is able to provide better discriminant ability in feature extraction for face biometrics. Science Publications 2010-07 Article PeerReviewed text/html en http://eprints.utm.my/26149/2/jcssp.2010.693.699 Shaikh Salleh, Sheikh Hussain and L. H., Chan and C. M., Ting (2010) Face biometrics based on principal component analysis and linear discriminant analysis. Journal of Computer Science, 6 (7). 693 -699. ISSN 1549-3636 DOI:10.3844/jcssp.2010.693.699 |
| spellingShingle | QA75 Electronic computers. Computer science Shaikh Salleh, Sheikh Hussain L. H., Chan C. M., Ting Face biometrics based on principal component analysis and linear discriminant analysis |
| title | Face biometrics based on principal component analysis and linear discriminant analysis |
| title_full | Face biometrics based on principal component analysis and linear discriminant analysis |
| title_fullStr | Face biometrics based on principal component analysis and linear discriminant analysis |
| title_full_unstemmed | Face biometrics based on principal component analysis and linear discriminant analysis |
| title_short | Face biometrics based on principal component analysis and linear discriminant analysis |
| title_sort | face biometrics based on principal component analysis and linear discriminant analysis |
| topic | QA75 Electronic computers. Computer science |
| url | http://eprints.utm.my/26149/2/jcssp.2010.693.699 |
| work_keys_str_mv | AT shaikhsallehsheikhhussain facebiometricsbasedonprincipalcomponentanalysisandlineardiscriminantanalysis AT lhchan facebiometricsbasedonprincipalcomponentanalysisandlineardiscriminantanalysis AT cmting facebiometricsbasedonprincipalcomponentanalysisandlineardiscriminantanalysis |