This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Markov Random Field Models for Unsupervised Segmentation of Textured Color Images
October 1995 (vol. 17 no. 10)
pp. 939-954

Abstract—We present an unsupervised segmentation algorithm which uses Markov random field models for color textures. These models characterize a texture in terms of spatial interaction within each color plane and interaction between different color planes. The models are used by a segmentation algorithm based on agglomerative hierarchical clustering. At the heart of agglomerative clustering is a stepwise optimal merging process that at each iteration maximizes a global performance functional based on the conditional pseudolikelihood of the image. A test for stopping the clustering is applied based on rapid changes in the pseudolikelihood. We provide experimental results that illustrate the advantages of using color texture models and that demonstrate the performance of the segmentation algorithm on color images of natural scenes. Most of the processing during segmentation is local making the algorithm amenable to high performance parallel implementation.

[1] R. Bajcsy,S.W. Lee,, and A. Leonardis,“Color image segmentation with detection of highlights and local illumination induced by interreflections,” Proc. Int’l Conf. Pattern Recognition, pp. 785-790,Atlantic City, N.J., 1990.
[2] J.-M. Beaulieu and M. Goldberg, "Hierarchy in Picture Segmentation: A Stepwise Optimization Approach," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 11, pp. 150-163, 1989.
[3] J. Besag,“Spatial interaction and the statistical analysis of lattice systems,” J. Royal Statistical Society B, vol. 36, pp. 192-236, 1974.
[4] R. Chellappa,S. Chatterjee,, and R. Bagdazian,“Texture synthesis and compression using Gaussian-Markov random field models,” IEEE Trans. Systems, Man, and Cybernetics, vol. 8, no. 2, pp. 298-303, Mar.-Apr. 1985.
[5] R. Chellapa,Y.-H. Hu,, and S.-Y. Kung,“On two-dimensional Markov spectral estimation,” IEEE Trans. Acoustics, Speech, and Signal Processing, vol. 31, no. 4, pp. 836-841, 1983.
[6] P.C. Chen and T. Pavlidis,“Image segmentation as an estimation problem,” Computer Graphics and Image Processing, vol. 12, pp. 153-172, 1980.
[7] F. Cohen,“Markov random fields for image modeling and analysis,” U. Desai, ed., Modeling and Application of Stochastic Processes, pp. 243-272.Boston: Kluwer Academic, 1986.
[8] F.S. Cohen and Z. Fan,“Maximum likelihood unsupervised textured image segmentation,” Computer Vision, Graphics, and Image Process, vol. 54, no. 3, pp. 239-251, May 1992.
[9] D. Cooper,“Stocahstic boudary estimation and object recognition,” Computer Graphics and Image Processing, pp. 326-355, Apr. 1980.
[10] M. Daily,“Color image segmentation using Markov random fields,” Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 304-312, 1989.
[11] A. Gagalowicz,S.D. Ma,, and C. Tournier-Lasserve,“Efficient models for color textures,” Proc. Eighth Int’l Conf. Pattern Recognition, pp. 412-414, 1986.
[12] S. Geman and D. Geman,“Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 6, no. 6, pp. 721-741, Nov. 1984.
[13] G. Healey,“Segmenting images using normalized color,” IEEE Trans. Systems, Man, and Cybernetics, vol. 22, pp. 64-73, Jan. 1992.
[14] G.E. Healey, S.A. Shafer, and L.B. Wolff, Physics-Based Vision Principles and Practice, Color. Boston: Jones and Bartlett, 1992.
[15] R.L. Kashyap and R. Chellappa,“Estimation and choice of neighbors in spatial-interaction models of images,” IEEE Trans. Information Theory, vol. 29, no. 1, pp. 60-72, Jan. 1983.
[16] R. Kashyap,R. Chellappa,, and N. Ahuja,“Decision rules for choice of neighbors in random field models of images,” Computer Graphics and Image Processing, vol. 15, no. 1, pp. 301-318, 1981.
[17] G. Klinker, S. Shafer, and T. Kanade, “A Physical Approach to Color Image Understanding,” Int'l J. Computer Vision, vol. 4, pp. 7-38, 1990.
[18] S. Lakshmanan and H. Derin, “Simultaneous Parameter Estimation and Segmentation of Gibbs Random Fields Using Simulated Annealing,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 11, pp. 799-813, 1989.
[19] W.E. Larimore,“Statistical interaction and statistical analysis of lattice systems,” Proc. IEEE, vol. 65, pp. 961-970, 1977.
[20] B.S. Manjunath and R. Chellappa, “Unsupervised Texture Segmentation Using Markov Random Field Models,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 13, pp. 478-482, 1991.
[21] Y. Ohta,T. Kanade,, and T. Sakai,“Color information for region segmentation,” Computer Graphics and Image Processing, vol. 13, pp. 222-241, 1980.
[22] D. Panjwani,“Segmentation of textured color images,” Master’s thesis, Dept. of Electrical and Computer Eng., Univ. of California, Irvine, 1993.
[23] D. Panjwani and G. Healey,“Selecting neighbors in random field models for color images,” Proc. First IEEE Int’l Conf. Image Processing,Austin, Tex., 1994.
[24] T. Pavlidis,Structural Pattern Recognition. Springer-Verlag, 1991.
[25] T. Poggio,E. Gamble,, and J. Little,“Parallel integration of vision modules,” Science, vol. 242, pp. 436-440, Oct. 1988.
[26] J. Scharcanski,J.K. Hovis,, and H.C. Shen,“Color texture representation using multiscale feature boundaries,” SPIE Vol. 1818 Visual Communications and Image Porcessing, pp. 156-165, 1992.
[27] J.F. Silverman and D.B. Cooper,“Bayesian clustering for unsupervised estimation of surface and texture models,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 10, no. 4, pp. 482-496, July 1988.
[28] J.W. Woods, “Two-Dimensional Discrete Markovian Fields,” IEEE Trans. Information Theory, vol. 18, pp. 232-240, 1972.
[29] W. Wright,“A Markov random field approach to data fusion and color segmentation,” Image and Vision Computing, vol. 7, pp. 144-150, 1989.
[30] Y. Yakimovsky,“Boundary and object detection in real world images,” J. ACM, vol. 23, pp. 599-618, 1976.

Index Terms:
Segmentation, color, texture, Markov random fields, machine vision, computer vision, color vision.
Citation:
Dileep Kumar Panjwani, Glenn Healey, "Markov Random Field Models for Unsupervised Segmentation of Textured Color Images," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 17, no. 10, pp. 939-954, Oct. 1995, doi:10.1109/34.464559
Usage of this product signifies your acceptance of the Terms of Use.