Kernel Null Space Marginal Fisher Analysis for Face Recognition

Abstract:

In the recent years, the feature extraction algorithms based on manifold learning, which attempt to project the original data into a lower dimensional feature space by preserving the local neighborhood structure, have drawn much attention. Among them, the Marginal Fisher Analysis (MFA) achieved high performance for face recognition. However, MFA suffers from the small sample size problems and is still a linear technique. This paper develops a new nonlinear feature extraction algorithm, called Kernel Null Space Marginal Fisher Analysis (KNSMFA). KNSMFA based on a new optimization criterion is presented, which means that all the discriminant vectors can be calculated in the null space of the within-class scatter. KNSMFA not only exploits the nonlinear features but also overcomes the small sample size problems. Experimental results on ORL database indicate that the proposed method achieves higher recognition rate than the MFA method and some existing kernel feature extraction algorithms.

• Periodical: Advanced Materials Research (Volumes 889-890)
• Pages: 1065-1068
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.889-890.1065
• Online since: February 2014
• Authors: Yu’e Lin*, Xing Zhu Liang, Hua Ping Zhou
• Keywords: Face Recognition, Marginal Fisher Analysis, Nonlinear Features, Small Sample Size Problems

References

[1] P Belhumeur, J Hespanha, D.J. Kriengman. Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection,. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 19(1997), p.711–720.

DOI: 10.1109/34.598228

[2] X F He, S C Yan, Y Hu, et al. Face recognition using Laplacianfaces,. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 27(2005), pp.328-340.

DOI: 10.1109/tpami.2005.55

[3] D Xu, S Yan, D Tao, et al. Marginal Fisher Analysis and Its Variants for Human Gait Recognition and Content- Based Image Retrieval,. IEEE Transactions on Image processing, Vol. 16(2007), pp.2811-2821.

DOI: 10.1109/tip.2007.906769

[4] B Scholkopf. A Smola. et al. Nonlinear component analysis as a kernel eigenvalue problem,. Neural Computer. vol. 9(1998), pp.1299-1319.

DOI: 10.1162/089976698300017467

[5] G Baudat, F Anouar. Generalized discriminant analysis using a kernel approach,. Neural Computer, vol. 12( 2000), pp.2385-2404.

DOI: 10.1162/089976600300014980