Multi-sensor fusion with Bayesian inference

  • Mark L. Williams
  • Richard C. Wilson
  • Edwin R. Hancock
Pattern Analysis
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1296)


This paper describes the development of a Bayesian framework for multiple graph matching. The study is motivated by the plethora of multi-sensor fusion problems which can be abstracted as multiple graph matching tasks. The study uses as its starting point the Bayesian consistency measure recently developed by Wilson and Hancock. Hitherto, the consistency measure has been used exclusively in the matching of graph-pairs. In the multiple graph matching study reported in this paper, we use the Bayesian framework to construct an inference matrix which can be used to gauge the mutual consistency of multiple graph-matches. The multiple graph-matching process is realised as an iterative discrete relaxation process which aims to maximise the elements of the inference matrix. We experiment with our multiple graph matching process using an application vehicle furnished by the matching of aerial imagery. Here we are concerned with the simultaneous fusion of optical, infra-red and synthetic aperture radar images in the presence of digital map data.


Synthetic Aperture Radar Synthetic Aperture Radar Image Graph Match Inference Process Correct Match 
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.
    Barrow H.G. and R.J. Popplestone, “Relational Descriptions in Picture Processing”, Machine Intelligence, 6, 1971.Google Scholar
  2. 2.
    Cross A.D.J., R.C. Wilson and E.R. Hancock, “Discrete Relaxation on a Boltzmann Machine”, Proceedings ICANN 95, pp. 491–496,1995.Google Scholar
  3. 3.
    Cross A.D.J., R.C.Wilson and E.R. Hancock, “Genetic Search for structural matching”, Proceedings ECCV96, LNCS 1064, pp. 514–525, 1996.Google Scholar
  4. 4.
    Cross A.D.J. and E.R.Hancock, “Relational Matching with stochastic optimisation” IEEE Computer Society International Symposium on Computer Vision, pp. 365–370, 1995.Google Scholar
  5. 5.
    Finch A.M., Wilson R.C. and Hancock E.R., “Softening Discrete Relaxation”, to appear in Neural Information Processing Systems 9, MIT Press 1997.Google Scholar
  6. 6.
    Gold S., A. Rangarajan and E. Mjolsness, “Learning with pre-knowledge: Clustering with point and graph-matching distance measures”, Neural Computation, 8, pp. 787–804, 1996.Google Scholar
  7. 7.
    Gold S. and A. Rangarajan, “A Graduated Assignment Algorithm for Graph Matching”, IEEE PAMI, 18, pp. 377–388, 1996.Google Scholar
  8. 8.
    Hathaway R.J., J.C. Bezdek, W. Pedrycz, “A parametric model for fusing heterogeneous fuzzy data”, IEEE Transactions on Fuzzy Systems, 4, pp. 270–281, 1996.CrossRefGoogle Scholar
  9. 9.
    Kittler J., W.J. Christmas and M.Petrou, “Probabilistic Relaxation for Matching Problems in Machine Vision”, Proceedings of the Fourth International Conference on Computer Vision, pp. 666–674, 1993.Google Scholar
  10. 10.
    Messmer B.T. and H. Bunke, " Efficient error-tolerant subgraph isomorphism detection”, Shape, Structure and Pattern Recognition, edited by D. Dori and A. Bruckstein, pp. 231–240, 1994.Google Scholar
  11. 11.
    Mjolsness E., G.Gindi and P. Anandan, “Optimisation in model matching and perceptual organisation”, Neural Computation, 1, pp. 218–219, 1989.Google Scholar
  12. 12.
    Sanfeliu A. and Fu K.S., “A Distance Measure Between Attributed Relational Graphs for Pattern Recognition”, IEEE SMC, 13, pp 353–362, 1983.Google Scholar
  13. 13.
    Shapiro L. and R.M. Haralick, “Structural Description and Inexact Matching”, IEEE PAMI, 3, pp 504–519, 1981.Google Scholar
  14. 14.
    Shapiro L. and R.M.Haralick, “A Metric for Comparing Relational Descriptions”, IEEE PAMI, 7, pp 90–94, 1985.Google Scholar
  15. 15.
    Simic P., “Constrained nets for graph matching and other quadratic assignment problems”, Neural Computation, 3, pp. 268–281, 1991.Google Scholar
  16. 16.
    Suganathan P.N., E.K. Teoh and D.P. Mital, “Pattern Recognition by Graph Matching using Potts MFT Networks”, Pattern Recognition, 28, pp. 997–1009, 1995.CrossRefGoogle Scholar
  17. 17.
    Tang Y.C. and C.S.G. Lee, “A Geometric Feature Relation Graph Formalism for Consistent Sensor Fusion”, IEEE SMC, 22, pp 115–129, 1992.Google Scholar
  18. 18.
    Tang Y.C. and C.S.G. Lee, “Optimal strategic recognition of objects based on candidate discriminating graph with coordinated sensors”, IEEE SMC, 22, pp. 647–661, 1992.Google Scholar
  19. 19.
    Wilson R.C., Evans A.N. and Hancock E.R., “Relational Matching by Discrete Relaxation”, Image and Vision Computing, 13, pp. 411–421, 1995.CrossRefGoogle Scholar
  20. 20.
    Wilson R.C and Hancock E.R., “Relational Matching with Dynamic Graph Structures”, Proceedings of the Fifth International Conference on Computer Vision, pp. 450–456, 1995.Google Scholar
  21. 21.
    Wilson R.C. and E.R. Hancock, “Gauging relational consistency and correcting structural errors”, IEEE Computer Society Computer Vision and Pattern Recognition Conference, pp. 47–54, 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Mark L. Williams
    • 1
  • Richard C. Wilson
    • 2
  • Edwin R. Hancock
    • 2
  1. 1.Defence Research AgencyMalvernUK
  2. 2.Department of Computer ScienceUniversity of YorkYorkUK

Personalised recommendations