Information-Theoretic Image Reconstruction and Segmentation from Noisy Projections

  • Gerhard Visser
  • David L. Dowe
  • Imants D. Svalbe
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5866)


The minimum message length (MML) principle for inductive inference has been successfully applied to image segmentation where the images are modelled by Markov random fields (MRF). We have extended this work to be capable of simultaneously reconstructing and segmenting images that have been observed only through noisy projections. The noise added to each projection depends on the classes of the pixels (material) that it passes through. The intended application is in low-dose (low-flux) X-ray computed tomography (CT) where irregular projections are used.


Image Segmentation Markov Random Field Inductive Inference Minimum Description Length Class Parameter 
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.
    Besag, J.E.: Spatial interaction and the statistical analysis of lattice systems. Journal of the Royal Statistical Society B36(2), 192–236 (1974)MathSciNetGoogle Scholar
  2. 2.
    Chaitin, G.J.: On the length of programs for computing finite binary sequences. Journal of the Association of Computing Machinery 13, 547–569 (1966)zbMATHMathSciNetGoogle Scholar
  3. 3.
    Comley, J.W., Dowe, D.L.: Minimum message length and generalized Bayesian nets with asymmetric languages. In: Grünwald, P., Pitt, M.A., Myung, I.J. (eds.) Advances in Minimum Description Length: Theory and Applications, pp. 265–294. MIT Press, Cambridge (2005)Google Scholar
  4. 4.
    Dalgleish, A.P., Dowe, D.L., Svalbe, I.D.: Tomographic reconstruction of images from noisy projections - a preliminary study. In: Orgun, M.A., Thornton, J. (eds.) AI 2007. LNCS (LNAI), vol. 4830, pp. 539–548. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  5. 5.
    Dowe, D.L.: Foreword re C. S. Wallace. Computer Journal 51(5), 523–560 (2008)CrossRefGoogle Scholar
  6. 6.
    Fayad, H., Guedon, J.P., Svalbe, I.D., Bizais, Y., Normand, N.: Applying mojette discrete radon transforms to classical tomographic data. In: Medical Imaging 2008: Physics of Medical Imaging. Proceedings of the SPIE, April 2008, vol. 6913, p. 69132S (2008)Google Scholar
  7. 7.
    Geman, S., Geman, D.: Stochastic relaxations, Gibbs distributions and the Bayesian restoration of images. IEEE Tran. on PAMI PAMI-6, 721–741 (1984)Google Scholar
  8. 8.
    Grimmett, G.R.: A theorem about random fields. Bull. London Math. Soc. 5, 81–84 (1973)zbMATHCrossRefMathSciNetGoogle Scholar
  9. 9.
    Kolmogorov, A.N.: Three approaches to the quantitative definition of information. Problems of Information Transmission 1, 1–17 (1965)Google Scholar
  10. 10.
    Needham, S.L., Dowe, D.L.: Message length as an effective Ockham’s razor in decision tree induction. In: Proc. 8th Int. Workshop of Artificial Intelligence and Statistics (AISTATS 2001), Key West, FL, pp. 253–260 (2001)Google Scholar
  11. 11.
    Rissanen, J.: Modeling by the shortest data description. Automatica 14, 465–471 (1978)zbMATHCrossRefGoogle Scholar
  12. 12.
    Solomonoff, R.J.: A formal theory of inductive inference. Information and Control 7, 1–22, 224–254 (1964)Google Scholar
  13. 13.
    Svalbe, I., van der Speck, D.: Reconstruction of tomographic images using analog projections and the digital radon transform. Linear Algebra and its Applications 339, 125–145 (2001)zbMATHCrossRefMathSciNetGoogle Scholar
  14. 14.
    Visser, G., Dowe, D.L.: Minimum message length clustering of spatially-correlated data with varying inter-class penalties. In: Proc. 6th IEEE International Conference on Computer and Information Science (ICIS 2007), Melbourne, Australia, July 2007, pp. 17–22 (2007)Google Scholar
  15. 15.
    Wallace, C.S.: An improved program for classification. In: Proceedings of the Nineteenth Australian Computer Science Conference (ACSC-9), Monash University, Australia, vol. 8, pp. 357–366 (1986)Google Scholar
  16. 16.
    Wallace, C.S.: False Oracles and SMML Estimators. In: Proc. Information, Statistics and Induction in Science conference (ISIS 1996), Was Tech Rept TR 89/128, Monash University, Australia, pp. 304–316. World Scientific, Singapore (1996)Google Scholar
  17. 17.
    Wallace, C.S.: Intrinsic classification of spatially correlated data. Computer Journal 41(8), 602–611 (1998)zbMATHCrossRefGoogle Scholar
  18. 18.
    Wallace, C.S.: Statistical and Inductive Inference by Minimum Message Length. Springer, Heidelberg (2005)zbMATHGoogle Scholar
  19. 19.
    Wallace, C.S., Boulton, D.M.: An information measure for classification. Computer Journal 11, 185–194 (1968)zbMATHGoogle Scholar
  20. 20.
    Wallace, C.S., Dowe, D.L.: Intrinsic classification by MML - the Snob program. In: Proc. 7th Australian Joint Conf. on Artificial Intelligence, pp. 37–44. World Scientific, Singapore (1994)Google Scholar
  21. 21.
    Wallace, C.S., Dowe, D.L.: Minimum message length and Kolmogorov complexity. Computer Journal 42(4), 270–283 (1999)zbMATHCrossRefGoogle Scholar
  22. 22.
    Wallace, C.S., Dowe, D.L.: MML clustering of multi-state, Poisson, von Mises circular and Gaussian distributions. Statistics and Computing 10, 73–83 (2000)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Gerhard Visser
    • 1
  • David L. Dowe
    • 1
  • Imants D. Svalbe
    • 1
  1. 1.Monash UniversityMelbourneAustralia

Personalised recommendations