research-article

Kernel-based nonlinear discriminant analysis for face recognition

Authors:

QingShan Liu,

Rui Huang,

HanQing Lu,

SongDe MaAuthors Info & Claims

Journal of Computer Science and Technology, Volume 18, Issue 6

Pages 788 - 795

https://doi.org/10.1007/BF02945468

Published: 01 November 2003 Publication History

Abstract

Linear subspace analysis methods have been successfully applied to extract features for face recognition. But they are inadequate to represent the complex and nonlinear variations of real face images, such as illumination, facial expression and pose variations, because of their linear properties. In this paper, a nonlinear subspace analysis method, Kernel-based Nonlinear Discriminant Analysis (KNDA), is presented for face recognition, which combines the nonlinear kernel trick with the linear subspace analysis method — Fisher Linear Discriminant Analysis (FLDA). First, the kernel trick is used to project the input data into an implicit feature space, then FLDA is performed in this feature space. Thus nonlinear discriminant features of the input data are yielded. In addition, in order to reduce the computational complexity, a geometry-based feature vectors selection scheme is adopted. Another similar nonlinear subspace analysis is Kernel-based Principal Component Analysis (KPCA), which combines the kernel trick with linear Principal Component Analysis (PCA). Experiments are performed with the polynomial kernel, and KNDA is compared with KPCA and FLDA. Extensive experimental results show that KNDA can give a higher recognition rate than KPCA and FLDA.

References

[1]

Zhao W, Chellappa R, Rosenfeld A, Phillips P J. Face Recognition: A Literature Survey.CS-Tech Report 4167, University of Maryland, 2000.

Google Scholar

[2]

Guo G D, Li Stan Z, Kapluk O. Face recognition by support vector machines. InProc. Int. Conf. Automatic Face and Gesture Recognition, Grenoble, France, 2000, pp. 196–201.

Google Scholar

[3]

Li S Z and Lu J W Face recognition using the nearest feature line method IEEE Trans. Neural Networks 1999 10 439-443

Crossref

Google Scholar

[4]

Liu C J, Weschsler H. A unified Bayesian framework for face recognition. InProc. Int. Conf. Image Processing, Brisbane, Australia, 1998, pp.151–155.

Google Scholar

[5]

Turk M and Pentland A Eigenfaces for recognition J. Cognitive Neuroscience 1991 3 1 71-86

Crossref

Google Scholar

[6]

Belhumeur P N, Hespanha J P, and Kriegman D J Eigen-face vs. fisherfaces: Recognition using class specific linear projection IEEE Trans. Pattern Analysis and Machine Intelligence 1997 19 7 711-720

Crossref

Google Scholar

[7]

Zhao W, Chellappa R, Phillips P J. Subspace Linear Discriminant Analysis for Face Recognition.Tech Report CAR-TR-914, Center for Antomation Research, University of Maryland, 1999.

Google Scholar

[8]

Chen L F, Liao H M, Lin J Cet al. A new IDA-based face recognition system which can solve the small sample size problem.Pattern Recognition, 2000, 33(10).

Google Scholar

[9]

Yu H and Yang J A direct LDA algorithm for high-dimensional data — With application to face recognition Pattern Recognition 2001 34 10 2067-2070

Crossref

Google Scholar

[10]

Huang R, Liu Q S, Lu H Q, Ma S D. Solving the small sample size problem of LDA. InProc. Int. Conf. Pattern Recognition, Quebec, Canada, 2002, 3: 29–32.

Google Scholar

[11]

Phillips P J, Moon H, Rizvi S, and Rauss P The Feret evaluation methodology for face recognition algorithms IEEE Trans. Pattern Analysis and Machine Intelligence 2000 22 10 1090-1104

Crossref

Google Scholar

[12]

M Stewart Bartlett, H Martin Lades, T Sejnowski. Independent component representations for face recognition. InProc. SPIE, San Jose, Canada, 1998, Vol. 3299, pp.528–539.

Google Scholar

[13]

Moghaddam B. Principal manifolds and Bayesian subspaces for visual recognition.Tech Report 99-35, Mitsubishi Electric Research Laboratory, 1999.

Google Scholar

[14]

Back K, Draper B A, Beveridge J Ret al. PCA vs ICA: A comparison on the FERET data set. http://www.cs.colostate.edu/evalfacerec/papers/cvprip02.pdf.

Google Scholar

[15]

Osuna E, Freund R, Girosi F. Support vector machines: Training and applications.Tech Report, AI Lab, MIT, 1997.

Google Scholar

[16]

Heisele B, Ho P, Poggio T. Face recognition with support vector machines: Global versus component-based approach. InProc. Int. Conf. Computer Vision, Vancouver, Canada, 2001, pp.688–673.

Google Scholar

[17]

Scholkopf B, Smola A, and Muller K R Nonlinear component analysis as a kernel eigenvalue problem Neural Computation 1998 10 1299-1319

Crossref

Google Scholar

[18]

Ming-Hsuan Yang, Narendra Ahuja, David Kriegman. Face recognition using kernel eigenfaces. InProc. Int. Conf. Image Processing, Vancouver, Canada, 2000, pp.37–40.

Google Scholar

[19]

Kin K I, Jung K, and Kim H J Face recognition using kernel principal component analysis IEEE Signal Processing Letters 2002 9 2 40-42

Crossref

Google Scholar

[20]

Liu Qingshan, Huang Rui, Lu Hanqinget al. Face recognition using kernel based fisher discriminant analysis. InProc. Int. Conf. Automatic Face and Gesture Recognition, Washington DC, USA, 2002, pp.197–201.

Google Scholar

[21]

Baudat G and Anouar F Generalized discriminant analysis using a kernel approach Neural Computation 2000 12 10 2385-2404

Crossref

Google Scholar

[22]

Mika S, Ratsch G, and Weston J Fisher discriminant analysis with kernels Neural Networks for Signal Processing 1999 IX 41-48

Google Scholar

[23]

Li Y M, Gong S G, Liddell H. Recognising trajectories of facial identities using kernel discriminant analysis. InProc. British Machine Vision Conference, 2001, pp.613–622.

Google Scholar

[24]

Wu Y, Huang T S, Toyama K. Self-supervised learning for object based on kernel discriminant-EM algorithm, InProc. Int. Conf. Compute. Vision, Vancouver, Canada, 2001, pp.275–280.

Google Scholar

[25]

Baudat G, Anouar F. Kernel-based methods and function approximation. InProc. Int. Conf. Neural Networks, Washington DC, July 15–19, 2001, pp.1244–1249.

Google Scholar

[26]

Phillips P J, Wechsler H, Huang J, and Rauss P The FERET database and evaluation procedure for face recognition algorithms Image and Vision Computing 1998 16 5 295-306

Crossref

Google Scholar

[27]

Yambor W S, Draper B A, Beveridge J R. Analyzing PCA-based face recognition algorithms: Eigen-vectors selection and distance measures.The 2nd Workshop on Empirical Evaluation in Computer Vision, http://www.cs.colostate.edu/~vision/papers/csueemcv.pdf, 2000.

Google Scholar

[28]

Freund Y and Schapire R E A decision-theoretic generalization of online learning and application to boosting Journal of Comp. & Sys. Sci 1997 55 1 119-139

Crossref

Google Scholar

Cited By

View all

Tongzhou ZYanli WHaihui WSheng GHongxian S(2009)Face recognition method by using large and representative datasetsProceedings of the 21st annual international conference on Chinese control and decision conference10.5555/1714810.1715108(5095-5098)Online publication date: 17-Jun-2009
https://dl.acm.org/doi/10.5555/1714810.1715108
Shen LJi ZBai L(2009)Tuning Kernel Parameters with Different Gabor Features for Face RecognitionProceedings of the 3rd International Conference on Intelligent Computing: Advanced Intelligent Computing Theories and Applications. With Aspects of Artificial Intelligence10.1007/978-3-540-74205-0_91(881-890)Online publication date: 17-Nov-2009
https://dl.acm.org/doi/10.1007/978-3-540-74205-0_91
Shen LBai LFairhurst M(2007)Gabor wavelets and General Discriminant Analysis for face identification and verificationImage and Vision Computing10.1016/j.imavis.2006.05.00225:5(553-563)Online publication date: 1-May-2007
https://dl.acm.org/doi/10.1016/j.imavis.2006.05.002

Index Terms

Kernel-based nonlinear discriminant analysis for face recognition
1. Computing methodologies
2. Mathematics of computing
  1. Mathematical analysis

Index terms have been assigned to the content through auto-classification.

Recommendations

Embedded Manifold-Based Kernel Fisher Discriminant Analysis for Face Recognition

Manifold learning algorithms mainly focus on discovering the intrinsic low-dimensional manifold embedded in the high-dimensional Euclidean space. Among them, locally linear embedding (LLE) is one of the most promising dimensionality reduction methods. ...
Face recognition using discriminant locality preserving projections based on maximum margin criterion

In this paper, we propose a new discriminant locality preserving projections based on maximum margin criterion (DLPP/MMC). DLPP/MMC seeks to maximize the difference, rather than the ratio, between the locality preserving between-class scatter and ...
Graph embedding discriminant analysis for face recognition

This paper develops a supervised discriminant technique, called graph embedding discriminant analysis (GEDA), for dimensionality reduction of high-dimensional data in small sample size problems. GEDA can be seen as a linear approximation of a ...

Reviews

Reviewer: Luminita State

In the field of face recognition, several nonlinear approaches have been proposed to refine extracted features while still keeping the computational complexity of the feature extraction process to within a reasonable level. This paper introduces kernel-based nonlinear discriminant analysis (KNDA), a nonlinear subspace analysis method created by combining a polynomial kernel technique, used to project the input data into an implicit feature space, with the performance of a Fisher linear discriminant analysis (FLDA) on the resulting feature space. A series of comments concerning other, similar approaches, and their reported efficiency in solving face recognition tasks, is presented in the introductory part of the paper. Next, the polynomial KNDA technique and the FLDA feature extraction method, expressed in terms of the between-class and the within-class scatter matrices, are presented. A scheme based on geometrical arguments is proposed in the third section, to reduce the computational complexity of the feature extraction process, yielding an iterative feature vectors selection algorithm. In the fourth section, the results of tests on three benchmarks (the ORL database, a subset of the FERET database, and the YALE database), performed using KNDA, kernel-based principal component analysis (KPCA), and FLDA techniques, and using in all cases the nearest neighbor classifier, are presented and commented on. The comparative analysis points out that KNDA+FLDA proves to be better than KPCA+FLDA in extracting features for face recognition. Also, the degree of the polynomial kernel, yielding different nonlinear space structures, strongly influences the performances of both the KNDA and KPCA methods. Finally, KNDA proved better than KPCA+FLDA for describing the complex variation of real face images, and feature vector selection is significantly better for computation reduction in KNDA, especially in the case of large databases. Online Computing Reviews Service

Access critical reviews of Computing literature here

Become a reviewer for Computing Reviews.

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image Journal of Computer Science and Technology

Journal of Computer Science and Technology Volume 18, Issue 6

Nov 2003

178 pages

ISSN:1000-9000

Issue’s Table of Contents

Publisher

Springer-Verlag

Berlin, Heidelberg

Publication History

Published: 01 November 2003

Revision received: 16 October 2002

Received: 24 June 2002

Author Tags

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

3
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 20 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

View all

Tongzhou ZYanli WHaihui WSheng GHongxian S(2009)Face recognition method by using large and representative datasetsProceedings of the 21st annual international conference on Chinese control and decision conference10.5555/1714810.1715108(5095-5098)Online publication date: 17-Jun-2009
https://dl.acm.org/doi/10.5555/1714810.1715108
Shen LJi ZBai L(2009)Tuning Kernel Parameters with Different Gabor Features for Face RecognitionProceedings of the 3rd International Conference on Intelligent Computing: Advanced Intelligent Computing Theories and Applications. With Aspects of Artificial Intelligence10.1007/978-3-540-74205-0_91(881-890)Online publication date: 17-Nov-2009
https://dl.acm.org/doi/10.1007/978-3-540-74205-0_91
Shen LBai LFairhurst M(2007)Gabor wavelets and General Discriminant Analysis for face identification and verificationImage and Vision Computing10.1016/j.imavis.2006.05.00225:5(553-563)Online publication date: 1-May-2007
https://dl.acm.org/doi/10.1016/j.imavis.2006.05.002

Abstract

References

Cited By

Index Terms

Recommendations

Embedded Manifold-Based Kernel Fisher Discriminant Analysis for Face Recognition

Face recognition using discriminant locality preserving projections based on maximum margin criterion

Graph embedding discriminant analysis for face recognition

Reviews

Access critical reviews of Computing literature here

Comments

Information

Published In

Publisher

Publication History

Author Tags

Qualifiers

Contributors

Other Metrics

Bibliometrics

Article Metrics

Other Metrics

Citations

Cited By

View options

Figures

Other

Share

Share this Publication link

Share on social media

Affiliations