Probabilistic Methods for Data Fusion

  • Ali Mohammad-Djafari
Part of the Fundamental Theories of Physics book series (FTPH, volume 98)


The main object of this paper is to show how we can use classical probabilistic methods such as Maximum Entropy (ME), maximum likelihood (ML) and/or Bayesian (BAYES) approaches to do microscopic and macroscopic data fusion. Actually ME can be used to assign a probability law to an unknown quantity when we have macroscopic data (expectations) on it. ML can be used to estimate the parameters of a probability law when we have microscopic data (direct observation). BAYES can be used to update a prior probability law when we have microscopic data through the likelihood. When we have both microscopic and macroscopic data we can use first ME to assign a prior and then use BAYES to update it to the posterior law thus doing the desired data fusion. However, in practical data fusion applications, we may still need some engineering feeling to propose realistic data fusion solutions. Some simple examples in sensor data fusion and image reconstruction using different kind of data are presented to illustrate these ideas.

key words

Data fusion Maximum entropy Maximum likelihood Bayesian data fusion EM algorithm 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    A. Mohammad-Djafari, Maximum Entropy and Linear Inverse Problems; A Short Review, pp. 253–264.Paris, France: Kluwer Academic Publishers, A. Mohammad-Djafari and G. Demoment ed., 1992.Google Scholar
  2. 2.
    A. Mohammad-Djafari, “Maximum ’entropie et problèmes inverses en im-agerie,” Traitement du Signal, pp. 87–116, 1994.Google Scholar
  3. 3.
    A. Mohammad-Djafari and J. Idier, Maximum Likelihood Estimation of the Lagrange Parameters of the Maximum Entropy Distributions, pp. 131–140.Seattle, USA: Kluwer Academic Publishers, C.R. Smith, G.J. Erikson and P.O. Neudorfer ed., 1991.Google Scholar
  4. 4.
    M. Miller and D. Snyder, “The role of likelihood and entropy in incomplete-data problems: Applications to estimating point-process intensities and toeplitz constrained covariances,” Proceedings of the IEEE, vol. 75, pp. 892–906, July 1987.CrossRefGoogle Scholar
  5. 5.
    S. Gautier, G. Le Besnerais, A. Mohammad-Djafaxi, and B. Lavayssière, Data fusion in the field of non destructive testing. Santa Fe, U.S.A.: Kluwer Academic Publishers, K. Hanson ed., 1995.Google Scholar
  6. 6.
    S. Gautier, G. Le Besnerais, A. Mohammad-Djafari, and B. Lavayssière, “Fusion de données radiographiques et ultrasonores, en vue ďune applicaion en contrôle non destructif,”(Clermont-Ferrand), Second international workshop on inverse problems in electromagnetism and acoustic, May 1995.Google Scholar
  7. 7.
    S. Gautier, Fusion de donnés gammagraphiques et ultrasonores. Application au contrôle non destructif. PhD thesis, Université de Paris-Sud, Orsay, septembre 1996.Google Scholar
  8. 8.
    S. Gautier, J. Idier, A. Mohammad-Djafari, and B. Lavayssière, “Fusion de données gammagraphiques et ultrasonores,” in GRETSI 97, (Grenoble, France), pp. 781–784,1997.Google Scholar
  9. 9.
    S. Gautier, B. Lavayssière, G. Le Besnerais, and A. Mohammad-Djafari, “L2+lp deconvolution and ultrasound imaging for non destructive evaluation,” in QLCAV 97, vol. 1, (Le Creusot), pp. 212–213, 1997.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 1998

Authors and Affiliations

  • Ali Mohammad-Djafari
    • 1
  1. 1.Laboratoire des Signaux et Systèmes (CNRS-SUPELEC-UPS)École Supérieure d’ÉlectricitéGif-sur-Yvette CedexFrance

Personalised recommendations