Energy Minimization by \(\alpha \)-Erosion for Supervised Texture Segmentation

  • Karl SkrettingEmail author
  • Kjersti Engan
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8814)


In this paper we improve image segmentation based on texture properties. The already good results achieved using learned dictionaries and Gaussian smoothing are improved by minimizing an energy function that has the form of a Potts model. The proposed \(\alpha \)-erosion method is a greedy method that essentially relabels the pixels one by one and is computationally very fast. It can be used in addition to, or instead of, Gaussian smoothing to regularize the label images in supervised texture segmentation problems. The proposed \(\alpha \)-erosion method achieves excellent results on a much used set of test images: on average we get 2.9 % wrongly classified pixels. Gaussian smoothing gives 10 % and the best results reported earlier give 4.5 %.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Besag, J.: On the statistical analysis of dirty pictures. Journal of the Royal Statistical Society. Series B (Methodological) 48(3), 259–302 (1986)MathSciNetzbMATHGoogle Scholar
  2. 2.
    Boykov, Y., Kolmogorov, V.: An experimental comparison of min-cut/max-flow algorithms for energy minimization in vision. IEEE Trans. Pattern Anal. Machine Intell. 26(9), 1124–1137 (2004)CrossRefGoogle Scholar
  3. 3.
    Boykov, Y., Veksler, O., Zabih, R.: Fast approximate energy minimization via graph cuts. IEEE Trans. Pattern Anal. Machine Intell. 23(11), 1222–1239 (2001)CrossRefGoogle Scholar
  4. 4.
    Delong, A., Gorelick, L., Veksler, O., Boykov, Y.: Minimizing energies with hierarchical costs. Int. J. Comput. Vision 100(1), 38–58 (2012)MathSciNetCrossRefzbMATHGoogle Scholar
  5. 5.
    Elad, M.: Sparse and Redundant Representations, from Theory to Applications in Signal and Image Processing. Springer, New York (2010)CrossRefzbMATHGoogle Scholar
  6. 6.
    Kohli, P., Ladický, L., Torr, P.H.S.: Robust higher order potentials for enforcing label consistency. Int. J. Comput. Vision 82(3), 302–324 (2009)CrossRefGoogle Scholar
  7. 7.
    Kolmogorov, V., Zabih, R.: What energy functions can be minimized via graph cuts. IEEE Trans. Pattern Anal. Machine Intell. 26, 147–159 (2004)CrossRefGoogle Scholar
  8. 8.
    Mäenpää, T., Pietikäinen, M., Ojala, T.: Texture classification by multi-predicate local binary pattern operators. In: Proc. ICPR (2000)Google Scholar
  9. 9.
    Mairal, J., Bach, F., Ponce, J., Sapiro, G., Zisserman, A.: Discriminative learned dictionaries for local image analysis. In: 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (June 2008)Google Scholar
  10. 10.
    Ojala, T., Mäenpää, T., Pietikäinen, M., Viertola, J., Kyllönen, J., Huovinen, S.: Outex - new framework for empirical evaluation of texture analysis algorithms. In: Proc. 16th Int. Conf. Pattern Recognition (2002)Google Scholar
  11. 11.
    Ojala, T., Pietikäinen, M., Mäenpää, T.: Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Trans. Pattern Anal. Machine Intell. 24(7), 971–987 (2002)CrossRefGoogle Scholar
  12. 12.
    Ojala, T., Valkealahti, K., Oja, E., Pietikäinen, M.: Texture discrimination with multidimensional distributions of signed gray-level differences. Pattern Recognition 34(3), 727–739 (2001)CrossRefzbMATHGoogle Scholar
  13. 13.
    Randen, T., Husøy, J.H.: Filtering for texture classification: A comparative study. IEEE Trans. Pattern Anal. Machine Intell. 21(4), 291–310 (1999)CrossRefGoogle Scholar
  14. 14.
    Skretting, K.: Sparse Signal Representation using Overlapping Frames. PhD thesis, NTNU Trondheim and Høgskolen i Stavanger (October 2002),
  15. 15.
    Skretting, K., Engan, K.: Recursive least squares dictionary learning algorithm. IEEE Trans. Signal Processing 58, 2121–2130 (2010)MathSciNetCrossRefGoogle Scholar
  16. 16.
    Skretting, K., Husøy, J.H.: Texture classification using sparse frame based representations. EURASIP Journal on Applied Signal Processing 2006, Article ID 52561 (2006)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  1. 1.Department of Electrical Engineering and Computer ScienceUniversity of StavangerStavangerNorway

Personalised recommendations