Non-bayesian Graph Matching without Explicit Compatibility Calculations

  • Barend Jacobus van Wyk
  • Michaël Antonie van Wyk
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2396)

Abstract

This paper introduces a novel algorithm for performing Attributed Graph Matching (AGM). A salient characteristic of the Interpolator-Based Kronecker Product Graph Matching (IBKPGM) algorithm is that it does not require the explicit calculation of compatibility values between vertices and edges, either using compatibility functions or probability distributions. No assumption is made about the adjacency structure of the graphs to be matched. The IBKPGM algorithm uses Reproducing Kernel Hilbert Space (RKHS) interpolator theory to obtain an unconstrained estimate to the Kronecker Match Matrix (KMM) from which a permutation sub-matrix is inferred.

Keywords

Reproduce Kernel Hilbert Space Graph Match Optimal Assignment Compatibility Function Pattern Recognition Letter 
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.

References

  1. 1.
    Gold, S., Rangarajan, A.: A Graduated Assignment Algorithm for Graph Matching, IEEE Trans. Patt. Anal. Machine Intell., Vol. 18(4) (1996)377–388CrossRefGoogle Scholar
  2. 2.
    You, M., Wong, K. C.: An Algorithm for Graph Optimal Isomorphism, Proc. ICPR, (1984) 316–319Google Scholar
  3. 3.
    Tsai, W.-H., Fu, K.-S.: Error-Correcting Isomorphisms of Attributed Relation Graphs for Pattern Recognition, IEEE Trans. Syst. Man Cybern., Vol. 9(1979)757–768MATHCrossRefGoogle Scholar
  4. 4.
    Tsai, W.-H., Fu, K.-S.: Subgraph Error-Correcting Isomorphisms for Syntactic Pattern Recognition, IEEE Trans. Systems, Man, Cybernetics, Vol. 13(1983)48–62MATHMathSciNetGoogle Scholar
  5. 5.
    Depiero, F., Trived, M., Serbin, S.: Graph Matching using a Direct Classification of Node Attendance, Pattern Recognition, Vol. 29(6) (1996) 1031–1048CrossRefGoogle Scholar
  6. 6.
    Eshera, M. A., Fu, K-S.: A Graph Distance Measure for Image Analysis, IEEE Trans. Syst. Man Cybern., Vol. 14(3)(1984)Google Scholar
  7. 7.
    Bunke, H., Messmer, B.: Recent Advances in Graph Matching, Int. J. Pattern Recognition Artificial Intell., Vol. 11(1)(1997)169–203CrossRefGoogle Scholar
  8. 8.
    Almohamad, H. A. L., Duffuaa, S. O.: A Linear Programming Approach for the Weighted Graph Matching Problem, IEEE Trans. Patt. Anal. Machine Intell., Vol. 15(5) (1993)522–525CrossRefGoogle Scholar
  9. 9.
    Almohamad, H. A. L.: Polynomial Transform for Matching Pairs of Weighted Graphs, Applied Mathematical Modelling, Vol. 15(4)(1991)216–222MATHCrossRefMathSciNetGoogle Scholar
  10. 10.
    Umeyama, S.: An Eigendecomposition Approach to Weighted Graph Matching Problems, IEEE Trans. Patt. Anal. Machine Intell., Vol. 10(5) (1988)695–703MATHCrossRefGoogle Scholar
  11. 11.
    Hummel, R. A., Zucker, S. W.: On the Foundations of Relaxation Labelling Processes, IEEE Trans. Patt. Anal. Machine Intell., Vol. 5(3)(1983)267–286MATHGoogle Scholar
  12. 12.
    Peleg, S.: A New Probabilistic Relaxation Scheme, IEEE Trans. Patt. Anal. Machine Intell., 2(4) (1980)362–369MATHCrossRefGoogle Scholar
  13. 13.
    Christmas, W. J., Kittler, J., Petrou, M.: Structural Matching in Computer Vision using Probabilistic Relaxation, IEEE Trans. Patt. Anal. Machine Intell., Vol. 17(8) (1995)749–764CrossRefGoogle Scholar
  14. 14.
    Finch, A. M., Wilson, R. C., Hancock, R.: Symbolic Matching with the EM Algorithm, Pattern Recognition, Vol. 31(11)(1998)1777–1790CrossRefGoogle Scholar
  15. 15.
    Williams, M. L., Wilson, R. C., Hancock, E. R.: Multiple Graph Matching with Bayesian Inference, Pattern Recognition Letters, Vol. 18(1997)1275–1281CrossRefGoogle Scholar
  16. 16.
    Cross, A. D. J., Hancock, E. R.: Graph Matching with a Dual Step EM Algorithm, IEEE Trans. Patt. Anal. Machine Intell., Vol. 20(11)(1998)1236–1253CrossRefGoogle Scholar
  17. 17.
    Wilson, R. C., Hancock, E. R.: A Bayesian Compatibility Model for Graph Matching, Pattern Recognition Letters, Vol. 17(1996)263–276CrossRefGoogle Scholar
  18. 18.
    Cross, A. D. J., Wilson, C., Hancock, E. R.: Inexact Matching Using Genetic Search, Pattern Recognition, Vol. 30(6) (1997)953–970CrossRefGoogle Scholar
  19. 19.
    Wilson, R. C., Hancock, E. R.: Structural Matching by Discrete Relaxation, IEEE Trans. Patt. Anal. Machine Intell., Vol. 19(8) (1997)634–648CrossRefGoogle Scholar
  20. 20.
    van Wyk, M. A., Durrani, T. S.: A Framework for Multi-Scale and Hybrid RKHS-Based Approximators, IEEE Trans. Signal Proc., Vol. 48(12) (2000)3559–3568CrossRefGoogle Scholar
  21. 21.
    van Wyk, M. A., Durrani, T. S., van Wyk, B. J.: A RKHS Interpolator-Based Graph Matching Algorithm, To appear in IEEE Trans. Patt. Anal. Machine Intell, July 2002Google Scholar
  22. 22.
    Van Wyk, M. A., Clark, J.: An Algorithm for Approximate Least-Squares Attributed Graph Matching, in Problems in Applied Mathematics and Computational Intelligence, N. Mastorakis (ed.), World Science and Engineering Society Press, (2001) 67–72Google Scholar
  23. 23.
    Luenberger, D. G.: Optimization by Vector Space Methods, New York, NY: John Wiley & Sons, 1969MATHGoogle Scholar
  24. 24.
    van Wyk, B. J., van Wyk, M. A., Virolleau, F.: The CGGM Algorithm and its DSP implementation, Proc. 3rd European DSP Conference on Education and Research, ESIEE-Paris, 20–21 September, 2000Google Scholar
  25. 25.
    Faugeras, O. D., Price, K. E.: Semantic Description of Aerial Images Using Stochastic Labeling, IEEE Trans. Patt. Anal. Machine Intell., Vol. 3(6)(1981)633–642Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Barend Jacobus van Wyk
    • 1
    • 2
  • Michaël Antonie van Wyk
    • 3
  1. 1.a division of DenelKentronCenturionSouth Africa
  2. 2.University of the WitwatersrandJohannesburgSouth Africa
  3. 3.Rand Afrikaans UniversityJohannesburgSouth Africa

Personalised recommendations