Markov Random Field Models for Unsupervised Segmentation of Textured Color Images

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]  John W. Woods,et al.  Two-dimensional discrete Markovian fields , 1972, IEEE Trans. Inf. Theory.

[2]  J. Besag.  Spatial Interaction and the Statistical Analysis of Lattice Systems , 1974 .

[3]  John W. Woods,et al.  Two-dimensional Markov spectral estimation , 1976, IEEE Trans. Inf. Theory.

[4]  Yoram Yakimovsky,et al.  Boundary and Object Detection in Real World Images , 1974, JACM.

[5]  Theodosios Pavlidis,et al.  Structural pattern recognition , 1977 .

[6]  W. Larimore,et al.  Statistical inference on stationary random fields , 1977, Proceedings of the IEEE.

[7]  Patrick C. Chen,et al.  Image segmentation as an estimation problem , 1979, 1979 18th IEEE Conference on Decision and Control including the Symposium on Adaptive Processes.

[8]  Patrick C. Chen,et al.  Image segmentation as an estimation problem , 1979, 1979 18th IEEE Conference on Decision and Control including the Symposium on Adaptive Processes.

[9]  T. Kanade,et al.  Color information for region segmentation , 1980 .

[10]  H. Elliott,et al.  Stochastic boundary estimation and object recognition , 1980 .

[11]  Ramalingam Chellappa,et al.  Decision rules for choice of neighbors in random field models of images , 1981 .

[12]  O.D. Faugeras,et al.  Applications of stochastic texture field models to image processing , 1981, Proceedings of the IEEE.

[13]  R. Chellappa,et al.  On two-dimensional Markov spectral estimation , 1983 .

[14]  Rama Chellappa,et al.  Estimation and choice of neighbors in spatial-interaction models of images , 1983, IEEE Trans. Inf. Theory.

[15]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Rama Chellappa,et al.  Texture synthesis and compression using Gaussian-Markov random field models , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[17]  Fernand S. Cohen,et al.  Markov random fields for image modelling and analysis , 1986 .

[18]  T Poggio,et al.  Parallel integration of vision modules. , 1988, Science.

[19]  David B. Cooper,et al.  Bayesian Clustering for Unsupervised Estimation of Surface and Texture Models , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  Michael J. Daily,et al.  Color image segmentation using Markov random fields , 1989, Proceedings CVPR '89: IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[21]  Morris Goldberg,et al.  Hierarchy in Picture Segmentation: A Stepwise Optimization Approach , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[22]  W. A. Wright.  A Markov random field approach to data fusion and colour segmentation , 1989, Image Vis. Comput..

[23]  Sridhar Lakshmanan,et al.  Simultaneous Parameter Estimation and Segmentation of Gibbs Random Fields Using Simulated Annealing , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[24]  R. Bajcsy,et al.  Color image segmentation with detection of highlights and local illumination induced by inter-reflections , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[25]  Huang Yumin,et al.  A PHYSICAL APPROACH TO COLOR IMAGE UNDERSTANDING , 1991 .

[26]  Rama Chellappa,et al.  Unsupervised Texture Segmentation Using Markov Random Field Models , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[27]  Zhigang Fan,et al.  Maximum likelihood unsupervised textured image segmentation , 1992, CVGIP Graph. Model. Image Process..

[28]  Helen C. Shen,et al.  Color texture representation using multiscale feature boundaries , 1992, Other Conferences.

[29]  Glenn Healey,et al.  Segmenting images using normalized color , 1992, IEEE Trans. Syst. Man Cybern..

[30]  Glenn Healey,et al.  Selecting neighbors in random field models for color images , 1994, Proceedings of 1st International Conference on Image Processing.

[31]  Toshio Uchiyama,et al.  Segmentation of Textured Color Images Using Multiscale Filtering , 1996 .