A New Approach to Unsupervised Markov Random Field-Based Segmentation of Mr Images

This paper describes a new approach to the unsupervised segmentation of images. A Markov random field model is used for prior label field modelling. Unlike conventional stochastic model-based approaches, each image class is not characterised by a parametric model. The algorithm compares the data in local windows around each pixel with the global distribution of data in each class using appropriate distance metrics. A novel method for determining the number of image classes is presented.

[1]  Donald Geman,et al.  Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images , 1984 .

[2]  T.F. Quatieri,et al.  Statistical model-based algorithms for image analysis , 1986, Proceedings of the IEEE.

[3]  Peter J. W. Rayner,et al.  Unsupervised image segmentation , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[4]  D R Haynor,et al.  Partial volume tissue classification of multichannel magnetic resonance images-a mixel model. , 1991, IEEE transactions on medical imaging.

[5]  Donald Geman,et al.  Boundary Detection by Constrained Optimization , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

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

[7]  Charles Kervrann,et al.  A Markov random field model-based approach to unsupervised texture segmentation using local and global spatial statistics , 1995, IEEE Trans. Image Process..

[8]  A. F. Smith,et al.  Statistical analysis of finite mixture distributions , 1986 .

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

[10]  Z. Liang,et al.  Tissue classification and segmentation of MR images , 1993, IEEE Engineering in Medicine and Biology Magazine.

[11]  H. Akaike.  A Bayesian analysis of the minimum AIC procedure , 1978 .

[12]  T Lei,et al.  Statistical approach to X-ray CT imaging and its applications in image analysis. II. A new stochastic model-based image segmentation technique for X-ray CT image , 1992, IEEE Trans. Medical Imaging.

[13]  M. Kasper.  graphics , 1991, Illustrating Mathematics.

[14]  Jun Zhang,et al.  Maximum-likelihood parameter estimation for unsupervised stochastic model-based image segmentation , 1994, IEEE Trans. Image Process..

[15]  Kenneth M. Hanson,et al.  Introduction to Bayesian image analysis , 1993 .

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

[17]  Thomas K. Pilgram,et al.  Validation of magnetic resonance imaging (MRI) multispectral tissue classification. , 1991, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[18]  R. Redner,et al.  Mixture densities, maximum likelihood, and the EM algorithm , 1984 .

[19]  Richard E. Wendell,et al.  Minimization of a Non-Separable Objective Function Subject to Disjoint Constraints , 1976, Oper. Res..

[20]  Wanqing Li,et al.  Unsupervised segmentation of dual-echo MR images by a sequentially learned Gaussian mixture model , 1995, Proceedings., International Conference on Image Processing.

[21]  J. Rissanen,et al.  Modeling By Shortest Data Description* , 1978, Autom..

[22]  Chee Sun Won,et al.  A parallel image segmentation algorithm using relaxation with varying neighborhoods and its mapping to array processors , 1987, Computer Vision Graphics and Image Processing.

[23]  G. B. Smith,et al.  Preface to S. Geman and D. Geman, “Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images” , 1987 .