Kernel feature extraction methods observed from the viewpoint of generating-kernels

Abstract

This paper introduces an idea of generating a kernel from an arbitrary function by embedding the training samples into the function. Based on this idea, we present two nonlinear feature extraction methods: generating kernel principal component analysis (GKPCA) and generating kernel Fisher discriminant (GKFD). These two methods are shown to be equivalent to the function-mapping-space PCA (FMS-PCA) and the function-mapping-space linear discriminant analysis (FMS-LDA) methods, respectively. This equivalence reveals that the generating kernel is actually determined by the corresponding function map. From the generating kernel point of view, we can classify the current kernel Fisher discriminant (KFD) algorithms into two categories: KPCA+ LDA based algorithms and straightforward KFD (SKFD) algorithms. The KPCA+ LDA based algorithms directly work on the given kernel and are not suitable for non-kernel functions, while the SKFD algorithms essentially work on the generating kernel from a given symmetric function and are therefore suitable for non-kernels as well as kernels. Finally, we outline the tensor-based feature extraction methods and discuss ways of extending tensor-based methods to their generating kernel versions.

Keywords

kernel methods / feature extraction / principal component analysis (PCA) / Fisher linear discriminant analysis (FLD or LDA) / tensor-based methods

References

[1] Vapnik V. The Nature of Statistical Learning Theory. New York: Springer, 1995

[2] Müller K R, Mika S, Rätsch G, Tsuda K, Schölkopf B. An introduction to kernel-based learning algorithms. IEEE Transactions on Neural Networks, 2001, 12(2): 181-201

[3] Schölkopf B, Smola A, Muller K R. Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation, 1998, 10(5): 1299-1319

[4] Mika S, Rätsch G, Weston J, Schölkopf B, Müller K R. Fisher discriminant analysis with kernels. In: Proceedings of IEEE International Workshop on Neural Networks for Signal Processing IX. 1999, 41-48

[5] Mika S, Rätsch G, Schölkopf B, Smola A, Weston J, Müller K R. Invariant feature extraction and classification in kernel spaces. Advances in Neural Information Processing Systems, 1999, 12: 526-532

[6] Baudat G, Anouar F. Generalized discriminant analysis using a kernel approach. Neural Computation, 2000, 12(10): 2385-2404

[7] Roth V, Steinhage V. Nonlinear discriminant analysis using kernel functions. In: Solla S A, Leen T K, Mueller K R, eds. Advances in Neural Information Processing Systems. 2000, 12: 568-574

[8] Mika S, Rätsch G, Weston J, Schölkopf B, Smola A, Müller K R. Constructing descriptive and discriminative non-linear features: Rayleigh coefficients in kernel feature spaces. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5): 623-628

[9] Yang M H. Kernel Eigenfaces vs. kernel Fisherfaces: face recognition using kernel methods. In: Proceedings of the 5th IEEE International Conference on Automatic Face and Gesture Recognition. 2002, 215-220

[10] Lu J, Plataniotis K N, Venetsanopoulos A N. Face recognition using kernel direct discriminant analysis algorithms. IEEE Transactions on Neural Networks, 2003, 14(1): 117-126

[11] Schölkopf B, Smola A. Learning with Kernels. Cambridge: MIT Press, 2002

[12] Shawe-Taylor J, Cristianini N. Kernel Methods for Pattern Analysis. Cambridge: Cambridge University Press, 2004

[13] Xu J, Zhang X, Li Y. Kernel MSE algorithm: a unified framework for KFD, LS-SVM, and KRR. In: Proceedings of the International Joint Conference on Neural Networks. 2001, 1486-1491

[14] Billings S A, Lee K L. Nonlinear fisher discriminant analysis using a minimum squared error cost function and the orthogonal least squares algorithm. Neural Networks, 2002, 15(2): 263-270

[15] Zheng W, Zhao L, Zou C. Foley-Sammon optimal discriminant vectors using kernel approach. IEEE Transactions on Neural Networks, 2005, 16(1): 1-9

[16] Yang J, Jin Z, Yang J Y, Zhang D, Frangi A F. Essence of kernel Fisher discriminant: KPCA plus LDA. Pattern Recognition, 2004, 37(10): 2097-2100

[17] Yang J, Frangi A F, Yang J Y, Zhang D, Jin Z. KPCA plus LDA: a complete kernel Fisher discriminant framework for feature extraction and recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(2): 230-244

[18] Xu Y, Zhang D, Jin Z, Li M, Yang J Y. A fast kernelbased nonlinear discriminant analysis for multi-class problems. Pattern Recognition, 2006, 39(6): 1026-1033

[19] Zhao J, Wang H, Ren H, Kee S C. LBP discriminant analysis for face verification. In: Proceeding of the 2005 IEEE Conference on Computer Vision and Pattern Recognition. 2005, 3: 167

[20] Liu C. Capitalize on dimensionality increasing techniques for improving face recognition grand challenge performance. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(5): 725-737

[21] Cevikalp H, Neamtu M, Wilkes M, Barkana A. Discriminative common vector method with kernels. IEEE Transactions on Neural Networks, 2006, 17(6): 1550-1565

[22] Bach F R, Jordan MI. Kernel independent component analysis. Journal of Machine Learning Research, 2002, 3(1): 1-48

[23] Yang J, Gao X, Zhang D, Yang J Y. Kernel ICA: an alternative formulation and its application to face recognition. Pattern Recognition, 2005, 38(10): 1784-1787

[24] Ma J. Function replacement vs. kernel trick. Neurocomputing, 2003, 50: 479-483

[25] Ma J, Theiler J, Perkins S. Two realizations of a general feature extraction framework. Pattern Recognition, 2004, 37(5): 875-887

[26] Lodhi H, Saunders C, Shawe-Taylor J, Cristianini N, Watkins C. Text classification using string kernels. Journal of Machine Learning Research, 2002, 2(2): 419-444

[27] Chen W S, Yuen P C, Huang J, Lai J. Wavelet kernel construction for kernel discriminant analysis on face recognition. In: Proceeding of the 2006 Conference on Computer Vision and Pattern Recognition Workshop. 2006, 47-52

[28] Schölkopf B. Support vector learning. Dissertation for the Doctoral Degree. Berlin: Berlin Technical University, 1997

[29] Camps-Valls G, Martin-Guerrero J, Rojo-Alvarez J, Soria-Olivas E. Fuzzy sigmoid kernel for support vector classifiers. Neurocomputing, 2004, 62: 501-506

[30] Ahonen T, Hadid A, Pietikäinen M. Face description with local binary patterns: application to face recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(12): 2037-2041

[31] Mangasarian L. Generalized support vector machines. In: Smola A J, Bartlett P, Schokopf B, Schuurmans D, eds. Advances in Large Margin Classifiers. 2000, 135-146

[32] Golub G H, Van Loan C F. Matrix Computations. 3rd ed. Baltimore: The Johns Hopkins University Press, 1996

[33] Swets D L,Weng J. Using discriminant eigenfeatures for image retrieval. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1996, 18(8): 831-836

[34] Belhumeur P N, Hespanha J P, Kriengman D J. Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(7): 711-720

[35] Yang J, Yang J Y. Why can LDA be performed in PCA transformed space? Pattern Recognition, 2003, 36(2): 563-566

[36] Liu C J, Wechsler H. Robust coding schemes for indexing and retrieval from large face databases. IEEE Transactions on Image Processing, 2000, 9(1): 132-137

[37] Turk M, Pentland A. Eigenfaces for recognition. Journal of Cognitive Neuroscience, 1991, 3(1): 71-86

[38] Liu K, Cheng Y Q, Yang J Y. Algebraic feature extraction for image recognition based on an optimal discriminant criterion. Pattern Recognition, 1993, 26(6): 903-911

[39] Yang J, Zhang D, Frangi A F, Yang J Y. Two-dimensional PCA: A new approach to appearance-based face representation and recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(1): 131-137

[40] Visani M, Garcia C, Laurent C. Comparing robustness of two-dimensional PCA and eigenfaces for face recognition. Lecture Notes in Computer Science, 2004, 3212: 717-724

[41] Zuo W M, Zhang D. K. Wang K. An assembled matrix distance metric for 2DPCA-based image recognition. Pattern Recognition Letters, 2006, 27(3): 210-216

[42] Zhang D Q, Zhou Z H. (2D)(2)PCA: Two-directional twodimensional PCA for efficient face representation and recognition. Neurocomputing, 2005, 69(1-3): 224-231

[43] Kong H,Wang L, Teoh E K, Li X, Wang J G, Venkateswarlu R. Generalized 2D principal component analysis for face image representation and recognition. Neural Networks, 2005, 18(5-6): 585-594

[44] Ye J. Generalized low rank approximations of matrices. Machine Learning, 2005, 61(1-3): 167-191

[45] Hyvärinen A, Oja E. Independent component analysis: algorithms and applications. Neural Networks, 2000, 13(4-5): 411-430

[46] He X, Yan S, Hu Y, Niyogi P, Zhang H J. Face recognition using Laplacianfaces. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(3): 328-340

[47] Ye J, Janardan R, Li Q. Two-dimensional linear discriminant analysis. Advances in Neural Information Processing Systems, 2004, 17: 1569-1576

[48] Li M, Yuan B. 2D-LDA: a novel statistical linear discriminant analysis for image matrix. Pattern Recognition Letters, 2005, 26(5): 527-532

[49] Xiong H, Swamy M, Ahmad M. Two-dimensional FLD for face recognition. Pattern Recognition, 2005, 38(7): 1121-1124

[50] Yang J, Zhang D, Yong X, Yang J. Two-dimensional linear discriminant transform for face recognition. Pattern Recognition, 2005, 38(7): 1125-1129

[51] Zuo W, Zhang D, Yang J, Wang K. BDPCA plus LDA: A novel fast feature extraction technique for face recognition. IEEE Transactions on Systems, Man, and Cybernetics, Part B, 2006, 36(4): 946-953

[52] Zou C, Sun N, Ji Z, Zhao L. 2DCCA: a novel method for small sample size face recognition. In: Proceedings of IEEE Workshop on Applications of Computer Vision. 2007, 43

[53] Lee S H, Choi S. Two-dimensional canonical correlation analysis. IEEE Signal Processing Letters, 2007, 14(10): 735-738

[54] Gao Q, Zhang L, Zhang D, Xu H. Independent components extraction from image matrix. Pattern Recognition Letters, 2010, 31(3): 171-178

[55] Chen S B, Zhao H F, Kong M, Luo B. 2d-lpp: a twodimensional extension of locality preserving projections. Neurocomputing, 2007, 70(4-6): 912-921

[56] Hu D W, Feng G Y, Zhou Z T. Two-dimensional locality preserving projections (2DLPP) with its application to palmprint recognition. Pattern Recognition, 2007, 40(1): 339-342

[57] Xu D, Yan S, Zhang L, Lin S, Zhang H J, Huang T S. Reconstruction and recognition of tensor-based objects with concurrent subspaces analysis. IEEE Transactions on Circuits and Systems for Video Technology, 2008, 18(1): 36-47

[58] Wang H, Ahuja N. A tensor approximation approach to dimensionality reduction. Journal of Computer Vision, 2008, 76(3): 217-229

[59] Tao D, Li X, Wu X, Maybank S J. General tensor discriminant analysis and gabor features for gait recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(10): 1700-1715

[60] Lu H, Plataniotis K N, Venetsanopoulos A N. MPCA: multilinear principal component analysis of tensor objects. IEEE Transactions on Neural Networks, 2008, 19(1): 18-39

[61] Zhang L, Gao Q, Zhang D. Directional independent component analysis with tensor representation. In: Proceedings of Computer Vision and Pattern Recognition. 2008, 1-7

[62] Sun N, Wang H, Ji Z, Zou C, Zhao L. An efficient algorithm for kernel two-dimensional principal component analysis. Neural Computing & Applications, 2008, 17(1): 59-64

[63] Liu J, Chen S, Zhou Z H, Tan X. Generalized low-rank approximations of matrices revisited. IEEE Transactions on Neural Networks, 2010, 21(4): 621-632

RIGHTS & PERMISSIONS

2014 Higher Education Press and Springer-Verlag Berlin Heidelberg