Advertisement

Advances in the statistical methodology for the selection of image descriptors for visual pattern representation and classification

  • Pavel Pudil
  • Jana Novovičová
  • Francesc Ferri
  • Josef Kittler
Posters
Part of the Lecture Notes in Computer Science book series (LNCS, volume 970)

Abstract

Recent advances in the statistical methodology for selecting optimal subsets of features (image descriptors) for visual pattern representation and classification are presented. The paper attempts to provide a guideline about which approach to choose with respect to the a priori knowledge of the problem. Two basic approaches are reviewed and the conditions under which they should be used are specified. References to more detailed material about each one of the methods are given and experimental results supporting the main conclusions are briefly outlined.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    A. P. Dempster, N. M. Laird, D. B. Rubin. Maximum likelihood from incomplete data via the EM algorithm. J. Roy. Stat. Soc., 39:1–38, 1977.Google Scholar
  2. 2.
    P.Devijver, J.Kittler. Pattern Recognition: A Statistical Approach. Prentice, 1982.Google Scholar
  3. 3.
    F. J. Ferri, P. Pudil, M. Hatef, J. Kittler. Comparative study of techniques for large-scale feature selection. In E. S. Gelsema, L. N. Kanal, eds, Pattern Recognition in Practice IV, pages 403–413. Elsevier, 1994.Google Scholar
  4. 4.
    L. Kanal. Patterns in pattern recognition: 1968–1974. IEEE Transactions on Information Theory, IT-20(6):697–722, November 1974.Google Scholar
  5. 5.
    J. Kittler. Feature selection and extraction. Handbook of Pattern Recognition and Image Processing, T. Y. Young, K. S. Fu, Eds., pp. 60–81, Academic Press, 1986.Google Scholar
  6. 6.
    M. Nadler, E.P. Smith. Pattern Recognition Engineering. John Wiley, 1992.Google Scholar
  7. 7.
    J. Novovičová, P. Pudil, J. Kittler. Feature selection based on divergence for empirical class densities. 9th Scandinavian Conf. on Image Analysis, Uppsala, 1995.Google Scholar
  8. 8.
    P. Pudil, S. Bláha, J. Novovičová. PREDITAS — software package for solving pattern recognition and diagonostic problems. Proc. BPRA 4th Intl. Conf. on Pattern Recognition, Cambridge (J. Kittler, ed.) Springer-Verlag, pp. 146–152, 1988.Google Scholar
  9. 9.
    P. Pudil, F. Ferri, J. Novovičová, J. Kittler. Floating search methods for feature selection with nonmonotonic criterion functions. Proceedings of the IEEE Intl. Conf. on Pattern Recognition, pp. 279–283, 1994.Google Scholar
  10. 10.
    P. Pudil, J. Novovičová, J.Kittler. Automatic machine learning of decision rule for classification problems in image analysis. In Proceedings of BMVC '93 — the 4th British Machine Vision Conference, 1993.Google Scholar
  11. 11.
    P. Pudil, J. Novovičová, J. Kittler. Simultaneous learning of decision rules and important attributes for classification problems in image analysis. Image and Vision Computing, 12(3):193–198, 1994.Google Scholar
  12. 12.
    P. Pudil, J. Novovičová, J. Kittler. Floating search methods in feature selection. Pattern Recognition Letters, (submitted).Google Scholar
  13. 13.
    W. Siedlecki, J. Sklansky. On automatic feature selection. International Journal of Pattern Recognition and Artificial Intelligence, 2(2):197–220, June 1988.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Pavel Pudil
    • 1
  • Jana Novovičová
    • 1
  • Francesc Ferri
    • 2
  • Josef Kittler
    • 3
  1. 1.Institute of Information Theory and AutomationAcademy of Sciences of the Czech RepublicPrague 8Czech Republic
  2. 2.Dept. d'Informatica i ElectronicaUniversitat de ValenciaBurjassot (Valencia)Spain
  3. 3.Dept. of Electronic and Electrical EngineeringUniversity of SurreyEngland

Personalised recommendations