Abstract
In this paper, we propose a new kernel discriminant analysis called kernel relevance weighted discriminant analysis (KRWDA) which has several interesting characteristics. First, it can effectively deal with the small sample size problem by using a QR decomposition on scatter matrices. Second, by incorporating a weighting function into discriminant criterion, it overcomes overemphasis on well-separated classes and hence can work under more realistic situations. Finally, using kernel theory, it handle non linearity efficiently. In order to improve performance of the proposed algorithm, we introduce two novel kernel functions and compare them with some commonly used kernels on face recognition field. We have performed multiple face recognition experiments to compare KRWDA with other dimensionality reduction methods showing that KRWDA consistently gives the best results.
Similar content being viewed by others
Notes
Kernels parameters: polynomial (d = 2), fractional power polynomial (d = 0.4), Gaussian kernel (σ2 = 105), power distance and logarithmic kernels with (β = 2).
References
Zhao W, Chellappa R, Rosenfeld A, Phillips PJ (2000) Face recognition: a literature survey. CS-Tech Report-4167, University of Maryland
Fukunaga K (1990) Introduction to statistical pattern recognition. Academic Press, New York
Turk M, Pentland A (1991) Eigenfaces for recognition. J Cogn Neurosci 3(1):71–86
Belhumeur PN, Hespanha JP, Kriegman DJ (1997) Eigenfaces vs. fisherfaces: recognition using class specific linear projection. IEEE Trans PAMI 19(7):711–720
Swets DL, Weng J (1996) Using discriminant eigenfeatures for image retrieval. IEEE Trans Pattern Anal Mach Intell 18(8):831–836
Ye J, Li Q (2004) LDA/QR: an efficient and effective dimension reduction algorithm and its theoretical foundation. Pattern Recognition 37(4):851–854
Loog M, Duin RPW, Hacb-Umbach R (2001) Multiclass linear dimension reduction by weighted pairwise Fisher criteria. IEEE Trans PAMI 23(7):762–766
Lotlikar R, Kothari R (2000) Fractional-step dimensionality reduction. IEEE Trans Pattern Anal Mach Intell 22(6):623–627
Tang EK, Suganthan PN, Yao X, Qin AK (2005) Linear dimensionality reduction using relevance weighted LDA. Pattern recognition 38(4):485–493
Golub GH, Loan CFV (1996) Matrix computation, 3rd edn. The Johns Hopkins University Press, Baltimore
Baudat G, Anouar F (2000) Generalized discriminant analysis using a kernel approach. Neural Comput 12(10):2385–2404
Scholkopf B, Smola A, Muller KR (1998) Nonlinear component analysis as a Kernel eigenvalue problem. Neural Comput 10(5):1299–1319
Yang MH, Ahuja N, Kriegman D (2000) Face recognition using kernel eigenfaces. In: Proceedings of international conference on image processing, pp 37–40
Mika S, Ratsch G, Weston J (1999) Fisher discriminant analysis with kernels. Neural Netw Signal Process 9:41–48
Liu Q, Huang R, Lu H, Ma S (2002) Face recognition using kernel-based Fisher discriminant analysis. In: Proceedings of the 5th IEEE international conference on automatic face and gesture recognition, pp 205–211
Huang MH (2002) Kernel eigenfaces vs. Kernel Fisherfaces: face recognition using kernel methods. In: Proceedings of the 5th IEEE international conference on automatic face and gesture recognition, pp 215–220
Liu Q, Tang X, Lu H, Ma S (2006) Face recognition using kernel scatter-difference-based discriminant analysis. IEEE Trans Neural Netw 17:1081–1085
Liu C (2004) Gabor-based kernel PCA with fractional power polynomial models for face recognition. IEEE Trans Pattern Anal Mach Intell 26(5): 572–581
Scholkopf B, Smola A (2002) Learning with kernels: support vector machines, regularization, optimization and beyond. MIT Press, Cambridge
Georghiades A, Belhumeur P, Kriegman’s D (2001) From few to many: illumination cone models for face recognition under variable lighting and pose. IEEE Trans Pattern Anal Mach Intell 23(6):642–660
Acknowledgements
The authors wish to thank the anonymous reviewers, whose comments were a great help to raise the level of technical and formal correctness of this paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chougdali, K., Jedra, M. & Zahid, N. Kernel relevance weighted discriminant analysis for face recognition. Pattern Anal Applic 13, 213–221 (2010). https://doi.org/10.1007/s10044-009-0152-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10044-009-0152-3