Abstract
Kernel Optimal Unsupervised Discriminant Projection (KOUDP) is presented in this paper. The proposed method first maps the input data into a potentially much higher dimensional feature space by virtue of nonlinear kernel trick, and in such a way, nonlinear features is extracted by running UDP on the kernel matrix. The singularity problem of the non-local scatter matrix due to small sample size problem occurred in UDP is avoided. Experimental results on YALE database indicate that the proposed KOUDP method achieves higher recognition rate than the UDP method and other kernel-based learning algorithms.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Belhumeur PN, Hespanha JP, Kriengman DJ (1997) Eigenfaces vs. fisherfaces: recognition using class specific linear projection. IEEE Trans Pattern Anal Mach Intell 19(7):711–720
He XF, Yan SC, Hu Y et al (2005) Face recognition using Laplacianfaces. IEEE Trans Pattern Anal Mach Intell 27(3):328–340
He XF, Niyogi P (2004) Locality preserving projections. Proc Adv Neural Inf Process Sys 16:153–160 (MIT Press, Cambridge)
Zhang T, Huang K, Li X, Yang J, Tao D (2010) Discriminative orthogonal neighborhood-preserving projections for classification. IEEE Trans Syst Man Cybern Part B Cybern 40(1):253–263
Yang J, David Z, Yang JY, Ben N (2007) Globally maximizing, locally minimizing: unsupervised discriminant projection with applications to face and palm biometrics. IEEE Trans Pattern Anal Mach Intell 29(4):650–664
Scholkopf B, Smola A et al (1998) Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput 10(5):1299–1319
Cheng J, Liu QS, Lu HQ et al (2005) Supervised kernel locality preserving projections for face recognition. Neurocomputing 67:443–449
Li JB, Pan JS, Chu SC (2008) Kernel class-wise locality preserving projection. Inf Sci 178(7):1825–1835
Li JB, Pan JS, Chu SC (2011) Kernel self-optimized locality preserving discriminant analysis for feature extraction and recognition. Neurocomputing 74(17):3019–3027
Acknowledgments
This work was financially supported by Foundation of Youth teachers of Anhui University of Science and Technology(No. 2012QNZ10), Anhui Provincial Natural Science Foundation (No.1208085QF123) and Anhui Provincial Natural Science Foundation of Higher Education of China (No. KJ2012Z084).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Liang, X., Lin, Y., Li, J. (2013). Kernel Optimal Unsupervised Discriminant Projection and Its Application to Face Recognition. In: Yin, Z., Pan, L., Fang, X. (eds) Proceedings of The Eighth International Conference on Bio-Inspired Computing: Theories and Applications (BIC-TA), 2013. Advances in Intelligent Systems and Computing, vol 212. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-37502-6_11
Download citation
DOI: https://doi.org/10.1007/978-3-642-37502-6_11
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-37501-9
Online ISBN: 978-3-642-37502-6
eBook Packages: EngineeringEngineering (R0)