Feature Selection for Graph-Based Image Classifiers

  • Bertrand Le Saux
  • Horst Bunke
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3523)


The interpretation of natural scenes, generally so obvious and effortless for humans, still remains a challenge in computer vision. We propose in this article to design binary classifiers capable to recognise some generic image categories. Images are represented by graphs of regions and we define a graph edit distance to measure the dissimilarity between them. Furthermore a feature selection step is used to pick in the image the most meaningful regions for a given category and thus have a compact and appropriate graph representation.


Feature Selection Mutual Information Region Type Graph Match Edit Operation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Chapelle, O., Haffner, P., Vapnik, V.: SVMs for histogram-based image classification. IEEE Transactions on Neural Networks 10, 1055–1065 (1999)CrossRefGoogle Scholar
  2. 2.
    Li, J., Wang, J.Z.: Automatic linguistic indexing of pictures by a statistical modeling approach. IEEE Trans. PAMI 25, 1075–1088 (2003)Google Scholar
  3. 3.
    Beretti, S., Del Bimbo, A., Vicario, E.: Efficient matching and indexing of graph models in content-based retrieval. IEEE Trans. PAMI 23, 1089–1105 (2001)Google Scholar
  4. 4.
    Minka, T., Picard, R.: Interactive learning using a society of models. Pattern Recognition 30, 565–581 (1997)CrossRefGoogle Scholar
  5. 5.
    Comaniciu, D., Meer, P.: Robust analysis of feature spaces: Color image segmentation. In: Proceedings of CVPR, San Juan, Porto Rico, pp. 750–755 (1997)Google Scholar
  6. 6.
    Le Saux, B., Boujemaa, N.: Unsupervised robust clustering for image database categorization. In: Proceedings of ICPR, Quebec, Canada, pp. 259–262 (2002)Google Scholar
  7. 7.
    Bezdek, J.C.: Pattern Recognition with Fuzzy Objective Function Algorithms. Plenum Press, New-York (1981)zbMATHGoogle Scholar
  8. 8.
    Guyon, I., Elisseff, A.: An introduction to variable and feature selection. Journal of Machine Learning Research 3, 1157–1182 (2003)zbMATHCrossRefGoogle Scholar
  9. 9.
    Le Saux, B., Amato, G.: Image recognition for digital libraries. In: ACM Multimedia/International Workshop on Multimedia Information Retrieval (2004)Google Scholar
  10. 10.
    Battiti, R.: Using mutual information for selecting features in supervised neural network learning. Neural Networks 5, 537–550 (1994)CrossRefGoogle Scholar
  11. 11.
    Gray, R.M.: Entropy and Information Theory. Springer, New York (1990)zbMATHGoogle Scholar
  12. 12.
    Nilsson, N.J.: Principles of Artificial Intelligence. Tioga, Palo Alto (1980)zbMATHGoogle Scholar
  13. 13.
    Messmer, B.: Graph Matching Algorithms and Applications. PhD thesis, University of Bern (1995)Google Scholar
  14. 14.
    Wong, E.: Three-dimensional object recognition by attributed graphs. In: Bunke, H., Sanfeliu, A. (eds.) Synctatic and Structural Pattern Recognition - Theory and Applications, pp. 314–381 (1990)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Bertrand Le Saux
    • 1
  • Horst Bunke
    • 1
  1. 1.Institut für Informatik und Angewandte MathematikUniversity of BernBernSwitzerland

Personalised recommendations