Advertisement

Comparison of Combining Methods of Correlation Kernels in kPCA and kCCA for Texture Classification with Kansei Information

  • Yo Horikawa
  • Yujiro Ohnishi
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4522)

Abstract

The authors consider combining correlations of different orders in kernel principal component analysis (kPCA) and kernel canonical correlation analysis (kCCA) with the correlation kernels. We apply combining methods, e.g., the sums of the correlation kernels, Cartesian spaces of the principal components or the canonical variates and the voting of kPCAs and kCCAs output and compare their performance in the classification of texture images. Further, we apply Kansei information on the images obtained through questionnaires to the public to kCCA and evaluate its effectiveness.

Keywords

kernel method principal component analysis canonical correlation analysis correlation kernel combining classifiers Kansei information texture classification 

References

  1. 1.
    Schölkopf, B., et al.: Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation 10, 1299–1319 (1998)CrossRefGoogle Scholar
  2. 2.
    Kuss, M., Graepel, T.: The geometry of kernel canonical correlation analysis. Technical Report 108, Max Plank Institute for Biological Cybernetics (2002)Google Scholar
  3. 3.
    Melzer, T., et al.: Appearance models based on kernel canonical correlation analysis. Pattern Recognition 36, 1961–1971 (2003)CrossRefzbMATHGoogle Scholar
  4. 4.
    Hardoon, D.R., et al.: Canonical correlation analysis; an overview with application to learning methods. Technical Report CSD-TR-03-02, Dept. of Computer Science, University of London (2003)Google Scholar
  5. 5.
    Ruiz, A., López-de-Teruel, P.E.: Nonlinear kernel-based statistical pattern analysis. IEEE Trans. Neural Networks 12, 16–32 (2001)CrossRefGoogle Scholar
  6. 6.
    Müller, K.-R., et al.: An introduction to kernel-based learning algorithms. IEEE Trans. Neural Networks 12, 181–201 (2001)CrossRefGoogle Scholar
  7. 7.
    Popovici, V., Thiran, J.-P.: Higher order autocorrelations for pattern classification. In: Proc. IEEE 2001 International Conference on Image Processing (ICIP2001), pp. 724–727 (2001)Google Scholar
  8. 8.
    Popovici, V., Thiran, J.-P.: Pattern recognition using higher-order local autocorrelation coefficients. Pattern Recognition Letters 25, 1107–1113 (2004)CrossRefGoogle Scholar
  9. 9.
    McLaughlin, J.A., Raviv, J.: Nth-order autocorrelations in pattern recognition. Information and Control 12, 121–142 (1968)CrossRefzbMATHGoogle Scholar
  10. 10.
    Horikawa, Y.: Comparison of support vector machines with autocorrelation kernels for invariant texture classification. In: Proc. 17th International Conference on Pattern Recognition (ICPR 2004), vol. 1, pp. 660–663 (2004)Google Scholar
  11. 11.
    Horikawa, Y.: Use of autocorrelation kernels in kernel canonical correlation analysis for texture classification. In: Pal, N.R., Kasabov, N., Mudi, R.K., Pal, S., Parui, S.K. (eds.) ICONIP 2004. LNCS, vol. 3316, pp. 1235–1240. Springer, Heidelberg (2004)Google Scholar
  12. 12.
    Horikawa, Y.: Modification of correlation kernels in SVM, KPCA and KCCA in texture classification. In: Proc. 2005 International Joint Conference on Neural Networks (IJCNN 2005), vol. 2, pp. 2006–2011 (2005)Google Scholar
  13. 13.
    Kittler, J.K., et al.: On combining classifiers. IEEE Trans. Patt. Anal. Machine Intell. 20, 226–239 (1998)CrossRefGoogle Scholar
  14. 14.
    Jain, A.K., et al.: Statistical pattern recognition: A review. IEEE Trans. Patt. Anal. Machine Intell. 22, 4–37 (2000)CrossRefGoogle Scholar
  15. 15.
    Kurita, T., Kato, T.: Learning of personal visual impression for image database systems. In: Proc. the Second International Conference on Document Analysis and Recognition, pp. 547–552 (1993)Google Scholar
  16. 16.
    Brodatz, P.: Textures: A Photographic Album for Artists and Designers. Dover, New York (1966)Google Scholar
  17. 17.
    Chijiiwa, H.: Color Science (in Japanese), Fukumura, Japan (1983), J.Suzuki’s Website: http://www.teu.ac.jp/chiit/~jsuzuki/kanseiwr.html
  18. 18.

Copyright information

© Springer Berlin Heidelberg 2007

Authors and Affiliations

  • Yo Horikawa
    • 1
  • Yujiro Ohnishi
    • 2
  1. 1.Faculty of Engineering, Kagawa University, Takamatsu, 761-0396Japan
  2. 2.Ryobi Systems Corporation, Okayama, 700-8504Japan

Personalised recommendations