Synthesis of Representative Graphical Symbols by Computing Generalized Median Graph

  • Xiaoyi Jiang
  • Andreas Münger
  • Horst Bunke
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1941)


Median is a general concept of capturing the essential information of a given set of objects. In this work we adopt this concept to the problem of learning, or synthesis, of representative graphical symbols from given examples. Graphical symbols are represented by graphs. This way the learning task is transformed into that of computing the generalized median of a given set of graphs, which is a novel graph matching problem and solved by a genetic algorithm.


Genetic Algorithm Edit Distance Input Graph Graph Match Label Graph 
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.
    A. D. J. Cross, R. C. Wilson, and E. R. Hancock, Inexact graph matching using genetic search, Pattern Recognition, 30(6): 953–970, 1997. 185CrossRefGoogle Scholar
  2. 2.
    D. E. Goldberg, Genetic algorithms in search, optimization and machine learning, Addison-Wesley, 1989. 185Google Scholar
  3. 3.
    A. H. Habacha, Structural recognition of disturbed symbols using discrete relaxation, Proc. of 1st Int. Conf. on Document Analysis and Recognition, Saint Malo, France, 170–178, 1991. 183Google Scholar
  4. 4.
    A. Hutchinson, Algorithmic Learning, Oxford University Press, 1994. 183Google Scholar
  5. 5.
    X. Jiang, A. Münger, and H. Bunke, Computing the generalized median of a set of graphs, Proc. of 2nd ICPAR Workshop on Graph-based Representations, Haindorf, Austria, 115–124, 1999. 183, 185, 187Google Scholar
  6. 6.
    R. Kasturi and K. Tmobre (Eds.), Graphics Recognition: Methods and Applications, Springer-Verlag, 1996. 183Google Scholar
  7. 7.
    P. Kuner and B. Ueberreiter, Pattern recognition by graph matching: combinatorial versus continuous optimization, Int. Journal of Pattern Recognition and Artificial Intelligence, 2(3): 527–542, 1988. 183CrossRefGoogle Scholar
  8. 8.
    S. Lee, Recognizing hand-written electrical circuit symbols with attributed graph matching, in H. S. Baird, H. Bunke, and K. Yamamoto (Eds.), Structured Document Analysis, 340–358, Springer-Verlag, 1988. 183Google Scholar
  9. 9.
    S. W. Lee, J. H. Kim, and F. C. A. Groen, Translation-, rotation-, and scaleinvariant recognition of hand-drawn symbols in schematic diagrams, Int. Journal of Pattern Recognition and Artificial Intelligence, 4(1): 1–25, 1990. 183CrossRefGoogle Scholar
  10. 10.
    J. Lladós, G. Sánchez, and E. Martí, A string based method to recognize symbols and structural textures in architectural plans, Proc. of 2nd IAPR Workshop on Graphics Recognition, Nancy, France, 287–294, 1997. 183Google Scholar
  11. 11.
    D. Lopresti and J. Zhou, Using consensus sequence voting to correct OCR errors, Computer Vision and Image Understanding, 67(1): 39–47, 1997. 184CrossRefGoogle Scholar
  12. 12.
    B. T. Messmer and H. Bunke, A new algorithm for error-tolerant subgraph isomorphism detection, IEEE Trans. on Pattern Analysis and Machine Intelligence, 20(5): 493–504, 1998. 184CrossRefGoogle Scholar
  13. 13.
    A. Münger, Synthesis of prototype graphs from sample graphs, Diploma Thesis, University of Bern, 1998. (in German) 185Google Scholar
  14. 14.
    Y.-K. Wang, K.-C. Fan, and J.-T. Horng, Genetic-based search for error-correcting graph isomorphism, IEEE Trans. on Systems, Man and Cybernetics — Part B: Cybernetics, 27(4): 588–597, 1997. 185CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Xiaoyi Jiang
    • 1
  • Andreas Münger
    • 1
  • Horst Bunke
    • 1
  1. 1.Department of Computer ScienceUniversity of BernSwitzerland

Personalised recommendations