A multigrid approach to the Gibbsian classification of mammograms

Both formal and informal locally-adaptive cooling schedules have been suggested to improve the convergence rate of Gibbs (and Gibbs-like) classification algorithms. One strategy involves maintaining a global cooling schedule/visiting schedule which is turned on or off (or forcing extremal temperature values) at a site depending on the interiteration behaviour of the classifier. This (0,1)-valued behaviour of the cooling schedule is parameterized relative to the site. We give a preliminary discussion of a method of assigning such parameters based on a multigrid decomposition of the image. The application domain is mammography.

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

[2]  Gerhard Winkler,et al.  Image analysis, random fields and dynamic Monte Carlo methods: a mathematical introduction , 1995, Applications of mathematics.

[3]  Ian R. Greenshields,et al.  Asynchronous, parallel pseudo-Gibbs classification of the VF dataset , 1999, Proceedings 12th IEEE Symposium on Computer-Based Medical Systems (Cat. No.99CB36365).

[4]  Ian R. Greenshields.  Local moments, contractive IFS, and multiresolution decompositions of 3D imagery , 1994, Other Conferences.

[5]  T Daggett,et al.  A cluster computer system for the analysis and classification of massively large biomedical image data. , 1998, Computers in biology and medicine.

[6]  Chang Wen Chen,et al.  Cellular neural network architecture for Gibbs random-field-based image segmentation , 1998, J. Electronic Imaging.

[7]  Patrick Pérez,et al.  Unsupervised image classification with a hierarchical EM algorithm , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.