A novel unsupervised multispectral texture segmentation algorithm is introduced. The textured image segmentation is based on a causal adaptive regression model prediction for detecting different types of texture segments which are present at the image. Texture segments are detected in four mutually perpendicular directions in the image lattice. Every monospectral component is checked separately and single monospectral results are combined together. The predictor in each direction uses identical contextual information from the pixel's neighbourhood and can be evaluated using a robust recursive algorithm. The method suggested can be successfully applied also to other unsupervised image segmentation applications, e.g. range image segmentation, edge detection, etc.


Texture segmentation 


  1. 1.
    Broemeling,L.D.: Bayesian Analysis of Linear Models. New York, Dekker 1985Google Scholar
  2. 2.
    Haindl,M. “ Texture synthesis,” CWI Quarterly 4, 1991, no. 4, pp. 305–331.Google Scholar
  3. 3.
    Haindl,M., Šimberová, S.: A Multispectral Image Line Reconstruction method. In: Theory & Applications of Image Analysis. P. Johansen, S. Olsen Eds., World Scientific Publishing Co., Singapore, 1992Google Scholar
  4. 4.
    Haralick, R.M.: “Statistical and structural approaches to texture,” Proc. IEEE, vol. 67, 1979, pp. 786–804.Google Scholar
  5. 5.
    Ojala, T., Pietikainen, M.: “Unsupervised Texture Segmentation Using Feature Distributions,” Proc. of ICIAP'97, Firenze, pp. 311–316, 1997.Google Scholar
  6. 6.
    Sinha, S.S., Jain, R.: Handbook of Pattern Recognition and Image Processing. Wiley, New York, 1994Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Michal Haindl
    • 1
    • 2
  1. 1.Institute of Information Theory and Automation of the Czech Academy of SciencesCzech Republic
  2. 2.Center for Machine PerceptionCzech Technical UniversityPrahaCzech Republic

Personalised recommendations