Piecewise Multi-linear PDF Modelling, Using an ML Approach

  • Edgard Nyssen
  • Naren Naik
  • Bart Truyen
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2396)


This paper addresses the problem of estimating the model parameters of a piecewise multi-linear (PML) approximation to a probability density function (PDF). In an earlier paper, we already introduced the PML model and discussed its use for the purpose of designing Bayesian pattern classifiers. The estimation of the unknown model parameters was based on a least squares minimisation of the difference between the estimated PDF and the estimating PML function. Here, we show how a Maximum Likelihood (ML) approach can be used to estimate the unknown parameters and discuss the advantages of this approach. Subsequently, we briefly introduce its application in a new approach to histogram matching in digital subtraction radiography.


  1. 1.
    Edgard Nyssen, Luc Van Kempen, and Hichem Sahli. Pattern classification based on a piecewise multi-linear model for the class probability densities. In Advances in Pattern Recognition — proceedings SSPR2000 and SPR2000, pages 501–510, 2000.Google Scholar
  2. 2.
    Richard O. Duda and Peter E. Hart. Pattern Classification and Scene Analysis. John Wiley & Sons, 1973.Google Scholar
  3. 3.
    Julius T. Tou and Raphael C. Gonzales. Pattern Recognition Principles. Addison Wesley Publishing Company, 1979.Google Scholar
  4. 4.
    Robert Schalkoff. Pattern Recognition — Statistical, Structural and Neural Approaches. John Wiley & Sons, 1992.Google Scholar
  5. 5.
    Fang Sun, Shin’ichiro Omachi, and Hirotomo Aso. An algorithm for estimating mixture distributions of high dimensional vectors and its application to character recognition. In Proc. 11th Scandinavian Conference on Image Analysis, pages 267–274, 1999.Google Scholar
  6. 6.
    David L. Donoho, Iain M. Johnstone, Gérard Kerkyacharian, and Dominique Pi-card. Density estimation by wavelet thresholding. The Annals of Statistics, 24(2):508–539, 1996.zbMATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    A. P. Dempster, N. M. Laird, and D. B. Rubin. Maximum likelihood estimation from incomplete data via the EM algorithm. J. Royal Statistical Society, 39:1–38, 1977.zbMATHMathSciNetGoogle Scholar
  8. 8.
    U. E. Ruttimann, R. L. Webber, and E. Schmidt. A robust digital method for film contrast correction in substraction radiography. J. Periodont. Res., 21:486–495, 1986.CrossRefGoogle Scholar
  9. 9.
    U. E. Ruttimann, T. Okano, H.-G. Gröndahl, K. Gröndahl, and R. L. Webber. Exposure geometry and film contrast differences as bases for incomplete cancellation of irrelevant structures in dental subtraction radiography. Proc. SPIE, 314:372–377, 1981.Google Scholar
  10. 10.
    H. B. Bidasaria. A method for almost exact histogram matching for 2 digitized images. Computer Graphics and Image Processing, 34, 1986.Google Scholar
  11. 11.
    Rafael C. Gonzalez and Paul Wintz. Digital image processing. Addison-Wesley Publishing Company, Amsterdam, 1987.Google Scholar
  12. 12.
    R. Morandi and P. Constantini. Piecewise monotone quadratic histosplines. SIAM J. Stat. Comput., 10:397–406, 1989.zbMATHCrossRefGoogle Scholar
  13. 13.
    J. W. Schmidt, W. Heß, and T. Nordheim. Shape preserving histopolation using rational quadratic splines. Comput., 44:245–258, 1990.zbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Edgard Nyssen
    • 1
  • Naren Naik
    • 1
  • Bart Truyen
    • 1
  1. 1.Vakgroep Elektronica en Informatieverwerking (ETRO)Vrije Universiteit BrusselBrusselBelgium

Personalised recommendations