Advertisement

Unsupervised Color Image Segmentation Using Mean Shift and Deterministic Annealing EM

  • Wanhyun Cho
  • Jonghyun Park
  • Myungeun Lee
  • Soonyoung Park
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3483)

Abstract

We present an unsupervised segmentation algorithm combining the mean shift procedure and deterministic annealing expectation maximization (DAEM) called MS-DAEM algorithm. We use the mean shift procedure to determine the number of components in a mixture model and to detect their modes of each mixture component. Next, we have adopted the Gaussian mixture model (GMM) to represent the probability distribution of color feature vectors. A DAEM formula is used to estimate the parameters of the GMM which represents the multi-colored objects statistically. The experimental results show that the mean shift part of the proposed MS-DAEM algorithm is efficient to determine the number of components and initial modes of each component in mixture models. And also it shows that the DAEM part provides a global optimal solution for the parameter estimation in a mixture model and the natural color images are segmented efficiently by using the GMM with components estimated by MS-DAEM algorithm.

Keywords

Color Image Gaussian Mixture Model Kernel Density Estimate Shift Vector Deterministic Annealing 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    McLachlan, G.J., Nguyen, S.K., Galloway, G.J., Wang, D.: Clustering of Magnetic Resonance Images. Technical Report, Department of Mathematics, University of Queensland (1998)Google Scholar
  2. 2.
    Permuter, H., Francos, J., Jermyn, I.H.: Gaussian mixture models of texture and colour for image database retrieval. In: Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing (2003)Google Scholar
  3. 3.
    Delignon, Y., Marzouki, A., Pieczynski, W.: Estimation of Generalized Mixtures and Its application in Image Segmentation. IEEE Transactions on Image Processing 6(10), 1364–1375 (1997)CrossRefGoogle Scholar
  4. 4.
    Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. Journal of Royal Statistical Society B 39, 1–39 (1977)zbMATHMathSciNetGoogle Scholar
  5. 5.
    Hofmann, T., Buhman, J.M.: Pairwise data clustering by deterministic annealing. IEEE Transactions on PAMI 19(1), 1–13 (1998)Google Scholar
  6. 6.
    Ueda, N., Nakano, R.: Deterministic annealing EM algorithm. Neural Networks 11, 271–282 (1998)CrossRefGoogle Scholar
  7. 7.
    Comaniciu, D., Meer, P.: Mean Shift: A Robust Approach Toward Feature Space Analysis. IEEE Transactions on PAMI 24(5), 1–17 (2002)Google Scholar
  8. 8.
    Cheng, Y.: Mean Shift, Mode Seeking, And Clustering. IEEE Transactions on PAMI 17(8), 790–799 (1995)Google Scholar
  9. 9.
    Kam, A.H., Fitzgerald, W.J.: Unsupervised Multiscale Image Segmentation. In: Proc. 3rd International Conference on Computer Vision, Pattern Recognition and Image Processing (CVPRIP 2000), Atlantic City, New Jersey, USA, vol. I, pp. 54–57 (2000)Google Scholar
  10. 10.
    Wang, H., Suter, D.: A Novel Robust Method for Large Numbers of Gross Errors. In: Seventh International Conference on Control, Automation, Robotics And Vision, pp. 326–331 (2002)Google Scholar
  11. 11.
    Park, J., Cho, W., Park, S.: Color Image Segmentation Using a Gaussian Mixture Model and a Mean field Annealing EM. IEICE Transactions on Information and Systems E86-D(10), 2240–2248 (2003)MathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Wanhyun Cho
    • 1
  • Jonghyun Park
    • 2
  • Myungeun Lee
    • 3
  • Soonyoung Park
    • 3
  1. 1.Department of StatisticsChonnam National UniversityChonnamSouth Korea
  2. 2.Institute for Robotics and Intelligent SystemsUniversity of Southern CaliforniaLos AngelesUSA
  3. 3.Department of Electronics EngineeringMokpo National UniversityChonnamSouth Korea

Personalised recommendations