Unsupervised Color Image Segmentation Using Compound Markov Random Field Model

  • Sucheta Panda
  • P. K. Nanda
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5909)


In this paper, we propose an unsupervised color image segmentation scheme using homotopy continuation method and Compound Markov Random Field (CMRF) model. The proposed scheme is recursive in nature where model parameter estimation and the image label estimation are alternated. Ohta (I 1, I 2, I 3) model is used as the color model for image segmentation and we propose a compound MRF model taking care of intra-color and inter-color plane interactions. The CMRF model parameters are estimated using Maximum Conditional Pseudo Likelihood (MCPL) criterion and the MCPL estimates are obtained using homotopy continuation method. The image label estimation is formulated using Maximum a Posteriori criterion and the MAP estimates are obtained using hybrid algorithm. In the context of misclassification error, the proposed unsupervised scheme with CMRF model exhibited improved segmentation accuracy as compared to MRF model and Kato’s method.


Color Image Color Model Segmentation Simulated Annealing MRF model 


  1. 1.
    Cheng, H.D., Jiang, X.H., Sun, Y., Wang, J.: Color Image Segmentation: Advances and prospects. Pattern Recog. 34, 2259–2281 (2001)zbMATHCrossRefGoogle Scholar
  2. 2.
    Ohta, Y.I., Kanade, T., Sakai, T.: Color information for region segmentation. Comp. Grap. Image. Process. 62, 222–241 (1980)CrossRefGoogle Scholar
  3. 3.
    Geman, S., Geman, D.: Stochastic relaxation,Gibbs distributions and the Bayesian restoration of images. IEEE Trans. on PAMI 6, 721–741 (1984)zbMATHGoogle Scholar
  4. 4.
    Zhang, J., Modestino, J.W.: A Model-Fitting Approach to Cluster Validation with Application to Stochastic Model-based Image Segmentation. IEEE Trans. on PAMI 12(10), 1009–1017 (1990)Google Scholar
  5. 5.
    Kato, Z., Pong, T.C., Lee, J.C.M.: Color image segmentation and parameter estimation in a markovian framework. Pattern Recognition Letters 22, 309–321 (2001)zbMATHCrossRefGoogle Scholar
  6. 6.
    Chow, N., Mallet-Paret, J., Yorke, J.A.: Finding zeros of maps: homotopy methods that are constructive with probability one. Math. Computation 32(143), 887–899 (1978)zbMATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    Stonick, V.L., Alexander, S.T.: A Relationship between recursive least square update and homotopy continuation methods. IEEE Trans. Signal Processing 39(2), 530–532 (1991)CrossRefGoogle Scholar
  8. 8.
    Wendell, R.E., Horter Jr., A.P.: Minimization of a non-separable objective function subject to disjoint constraints. Operations Research 24(4), 643–657 (1976)zbMATHCrossRefMathSciNetGoogle Scholar
  9. 9.
    Sucheta, P., Nanda, P.K.: Color Image using Constrained Compound Markov Random Field Model and Homotopy Continuation Method. In: Proc. of the first International Conference on Distributed Frameworks and Applications, Universiti Sains Malaysia, Penang, Malaysia, pp. 151–158 (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Sucheta Panda
    • 1
  • P. K. Nanda
    • 2
  1. 1.IPCV Lab., Department of Electrical EngineeringNational Institute of TechnologyRourkelaIndia
  2. 2.Department of Electronics and Telecommunication EngineeringC.V Raman College of EngineeringBhubaneswarIndia

Personalised recommendations