Unsupervised Segmentation of Markov Random Field Modeled Textured Images Using Selectionist Relaxation

Among the existing texture segmentation methods, those relying on Markov random fields have retained substantial interest and have proved to be very efficient in supervised mode. The use of Markov random fields in unsupervised mode is, however, hampered by the parameter estimation problem. The recent solutions proposed to overcome this difficulty rely on assumptions about the shapes of the textured regions or about the number of textures in the input image that may not be satisfied in practice. In this paper, an evolutionary approach, selectionist relaxation, is proposed as a solution to the problem of segmenting Markov random field modeled textures in unsupervised mode. In selectionist relaxation, the computation is distributed among a population of units that iteratively evolves according to simple and local evolutionary rules. A unit is an association between a label and a texture parameter vector. The units whose likelihood is high are allowed to spread over the image and to replace the units that receive lower support from the data. Consequently, some labels are growing while others are eliminated. Starting with an initial random population, this evolutionary process eventually results in a stable labelization of the image, which is taken as the segmentation. In this work, the generalized Ising model is used to represent textured data. Because of the awkward nature of the partition function in this model, a high-temperature approximation is introduced to allow the evaluation of unit likelihoods. Experimental results on images containing various synthetic and natural textures are reported.

[1]  Heinz Mühlenbein,et al.  Parallel Genetic Algorithms, Population Genetics, and Combinatorial Optimization , 1989, Parallelism, Learning, Evolution.

[2]  Moustafa M. Fahmy,et al.  Texture segmentation based on a hierarchical Markov random field model , 1992, Signal Process..

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

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

[5]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[6]  Kalyanmoy Deb,et al.  A Comparative Analysis of Selection Schemes Used in Genetic Algorithms , 1990, FOGA.

[7]  J. Laurie Snell,et al.  Markov Random Fields and Their Applications , 1980 .

[8]  H. Derin,et al.  Segmentation of textured images using Gibbs random fields , 1986 .

[9]  R. Collins.  Studies in artificial evolution , 1992 .

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

[11]  Gilbert Syswerda,et al.  Uniform Crossover in Genetic Algorithms , 1989, ICGA.

[12]  David B. Cooper,et al.  Simple Parallel Hierarchical and Relaxation Algorithms for Segmenting Noncausal Markovian Random Fields , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Reiko Tanese,et al.  Distributed Genetic Algorithms , 1989, ICGA.

[14]  Bernard Manderick,et al.  Fine-Grained Parallel Genetic Algorithms , 1989, ICGA.

[15]  Chaur-Chin Chen,et al.  Markov random fields for texture classification , 1993, Pattern Recognit. Lett..

[16]  R. Chellappa.  Two-Dimensional Discrete Gaussian Markov Random Field Models for Image Processing , 1989 .

[17]  Rama Chellappa,et al.  Classification of textures using Gaussian Markov random fields , 1985, IEEE Trans. Acoust. Speech Signal Process..

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

[19]  M. Hassner,et al.  The use of Markov Random Fields as models of texture , 1980 .

[20]  J. M. Hans du Buf,et al.  A review of recent texture segmentation and feature extraction techniques , 1993 .

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

[22]  Philippe Andrey,et al.  Unsupervised image segmentation using a distributed genetic algorithm , 1994, Pattern Recognit..

[23]  Bernard Manderick,et al.  A Massively Parallel Genetic Algorithm: Implementation and First Analysis , 1991, ICGA.

[24]  J. Besag.  On the Statistical Analysis of Dirty Pictures , 1986 .

[25]  Anil K. Jain,et al.  Random field models in image analysis , 1989 .

[26]  Paul Cohen,et al.  Gibbs Random Fields, Fuzzy Clustering, and the Unsupervised Segmentation of Textured Images , 1993, CVGIP Graph. Model. Image Process..

[27]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[28]  Anil K. Jain,et al.  Markov Random Field Texture Models , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[29]  Haluk Derin,et al.  Modeling and Segmentation of Noisy and Textured Images Using Gibbs Random Fields , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[30]  Rama Chellappa,et al.  Stochastic and deterministic networks for texture segmentation , 1990, IEEE Trans. Acoust. Speech Signal Process..

[31]  Georgy L. Gimel'farb,et al.  Probabilistic models of digital region maps based on Markov random fields with short- and long-range interaction , 1993, Pattern Recognit. Lett..

[32]  Dana S. Richards,et al.  Distributed genetic algorithms for the floorplan design problem , 1991, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[33]  Chee Sun Won,et al.  Unsupervised segmentation of noisy and textured images using Markov random fields , 1992, CVGIP Graph. Model. Image Process..

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