Compact Form of the Pseudo–inverse Matrix in the Approximation of a Star Graph Using the Conductance Electrical Model (CEM)

  • Manuel Igelmo
  • Alberto Sanfeliu
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7626)


The Conductance Electrical Model (CEM) transforms a graph into a circuit and can be use to do ”inexact graph isomorphism” as it was shown in [13]. In second stage of this process, we transform the circuit r eq in a star circuit, using the Moore–Penrose pseudo–inverse of a matrix for which there is a general formula that requires transpose, multiply and invert matrices with a time complexity of O(N 4), where N is the number of nodes of the graph. However, due to the special structure of the star transformation, we are able to exploit this special structure to compute the pseudo–inverse without using the general Moore–Penrose formula. We have developed a closed formula that can compute the elements of the pseudo–inverse without using that formula, that means without multiplying matrices neither doing the matrix inversion and that moreover can be computed in O(N 3). This method also eliminates the problems due to computer rounding and due to bad–conditioned problems in mathematical terms.


Inverse Matrix Matrix Inversion Graph Match Reference Node Graph Isomorphism 
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.


  1. 1.
    Bunke, H., Sanfeliu, A.: Syntactic and Structural Pattern Recognition – Theory and Applications. Series in Computer Science, vol. 7. World Scientific Publishing Co. Pte. Ltd, Singapore (1990)CrossRefzbMATHGoogle Scholar
  2. 2.
    Ullman, J.R.: An algorithm for subgraph isomorphism. Journal of the Association for Computing Machinery 23(1), 31–42 (1976)MathSciNetCrossRefzbMATHGoogle Scholar
  3. 3.
    Messmer, B.T., Bunke, H.: A new algorithm for error–tolerant subgraph isomorphism detection. IEEE Transactions on Pattern Analysys and Machine Intelligence 20(5), 493–504 (1998)CrossRefGoogle Scholar
  4. 4.
    Sanfeliu, A., Fu, K.S.: A distance measure between attributed relational graphs for pattern recognition. IEEE Trans. Syst. Man Cybern. SMC-13(3), 353–362 (1983)CrossRefzbMATHGoogle Scholar
  5. 5.
    Neuhaus, M., Bunke, H.: Automatic learning of cost function for graph edit distance. Information Sciences 177, 239–247 (2007)MathSciNetCrossRefzbMATHGoogle Scholar
  6. 6.
    Neuhaus, M., Bunke, H.: Edit distance–based kernel functions for structural pattern classification. Pattern Recognition 39(10), 1852–1863 (2006)CrossRefzbMATHGoogle Scholar
  7. 7.
    Sanfeliu, A., Alquézar, R., Andrade, J., Climent, J., Serratosa, F., Vergés, J.: Graph-based representations and techniques for image processing and image analysis. Pattern Recognition 35, 639–650 (2002)CrossRefzbMATHGoogle Scholar
  8. 8.
    Umeyama, S.: An Eigendecomposition approach to weighted graph matching problems. IEEE Trans. PAMI 10, 695–703 (1998)CrossRefGoogle Scholar
  9. 9.
    Robles–Kelly, A., Hancock, E.R.: An expectation–maximisation framework for segmentation and grouping. Image and Vision Computing 20, 725–738 (2002)CrossRefGoogle Scholar
  10. 10.
    Gori, M., Maggini, M., Sarti, L.: Exact and Approximate Graph Matching Using Random Walks. Pattern Anal. and Mach. Intelligence 27(7), 1100–1111 (2005)CrossRefGoogle Scholar
  11. 11.
    Cross, A.D.J., Wilson, R.C., Hancock, E.R.: Inexact Graph Matching Using Genetic Search. Pattern Recognition 30, 953–970 (1997)CrossRefGoogle Scholar
  12. 12.
    Wilson, R.C., Hancock, E.R.: Structural matching by discrete relaxation. Pattern Analysis and Machine Intelligence 19(6), 634–648 (1997)CrossRefGoogle Scholar
  13. 13.
    Igelmo, M., Sanfeliu, A., Ferrer, M.: A Conductance Electrical Model for Representing and Matching Weighted Undirected Graphs. In: Proceedings of the International Conference on Pattern Recognition (ICPR 2010), pp. 958–961 (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Manuel Igelmo
    • 1
  • Alberto Sanfeliu
    • 1
    • 2
  1. 1.Universitat Politècnica de Catalunya (UPC)Spain
  2. 2.Institut de Robòtica i Informàtica Industrial (UPC-CSIC)Spain

Personalised recommendations