Journal of Mathematical Imaging and Vision

, Volume 5, Issue 4, pp 277–286 | Cite as

Texture-based segmentation using markov random field models and approximate Bayesian estimators based on trees

  • Chi-Hsin Wu
  • Peter C. Doerschuk

Abstract

We describe segmentation based on textures using the label and image model of D. Gemanet al., “Boundary Detection by Constrained Optimization,”IEEE Trans. Pattern Analysis and Machine Intelligence, 12(7):609–628, July 1990. We replace their maximuma posteriori estimation criterion with a Bayesian estimator that minimizes the sum of the pixel misclassification probabilities. The new estimation goal allows the use of a different computational algorithm, which is deterministic rather than random, based on approximating lattices by trees. An example demonstrating an accurate segmentation of a collage of Brodatz textures is included.

Keywords

Image textures Image segmentation Markov random fields Kolmogorov-Smirnov statistic 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Julian Besag. On the statistical analysis of dirty pictures.J. Royal Stat. Soc. B, 48:259–302, 1986.Google Scholar
  2. 2.
    Donald Geman, Stuart Geman, Christine Graffigne, and Ping Dong. Boundary detection by constrained optimization.IEEE Trans. PAMI, 12(7):609–627, July 1990.Google Scholar
  3. 3.
    Stuart Geman and Donald Geman. Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images.IEEE Trans. PAMI, 6(6):721–741, November 1984.Google Scholar
  4. 4.
    Christine Graffigne.Experiments in Texture Analysis and Segmentation. PhD thesis, Brown University, Providence, RI, May 1987.Google Scholar
  5. 5.
    J. Marroquin, S. Mitter, and T. Poggio. Probabilistic solution of ill—posed problems in computational vision.J. Am. Stat. Assoc., 82(397):76–89, 1987.Google Scholar
  6. 6.
    J. M. Ortega and W. C. Rheinboldt.Iterative Solution of Nonlinear Equations in Several Variables. Academic Press, Inc., San Diego, 1970.Google Scholar
  7. 7.
    William H. Press, Brian P. Flannery, Saul A. Teukolsky, and William T. Vetterling.Numerical Recipes in C: The Art of Scientific Computing. Cambridge Univ. Press, Cambridge, 2 edition, 1992.Google Scholar
  8. 8.
    Chi-hsin Wu.Deterministic Parallelizable Solutions for Bayesian Markov Random Field Estimation Problems. PhD thesis, Purdue University, West Lafayette, IN, USA, May 1994.Google Scholar
  9. 9.
    Chi-hsin Wu and Peter C. Doerschuk. Tree approximations to Markov random fields.IEEE Trans. PAMI, 17(4):391–402, April 1995.Google Scholar

Copyright information

© Kluwer Academic Publishers 1995

Authors and Affiliations

  • Chi-Hsin Wu
    • 1
  • Peter C. Doerschuk
    • 2
  1. 1.Image Processing Department of the Opto-Electronics & Systems LaboratoryITRIHsinchuROC
  2. 2.School of Electrical EngineeringPurdue UniversityWest LafayetteUSA

Personalised recommendations