A Markovian Approach to Unsupervised Multidimensional Pattern Classification
This paper proposes a new method for core cluster detection prior to unsupervised automatic classification. Based upon a Markov random field model, this approach transforms the set of multidimensional observations into a normalised discret binary set, which represents the observable field. The field of classes is then represented by connex components corresponding to the cores, or prototypes, inside the samples. Classification results of artificially generated data are compared with results obtained by a classical clustering method.
Unable to display preview. Download preview PDF.
- P.A.DEVIJER and J.KITTLER (1982): Pattern Recognition. A statistical Approach. Englwood Cliff, NJ, Prentice-Hall International, 448.Google Scholar
- S. GEMAIN and P. GEMAIN (1984): Stochastic Relaxation, Gibbs distribution and the Baysien Restoration of Images. IEEE Trans on pattern anal. and Machine, Intell PAMI-6, 721–741.Google Scholar
- J. G. POSTAIRE and C. P. VASSEUR (1983): A Fast Algorithm for Non Parametric Probability Density Estimation. IEEE Trans on Pattern Anal. Machine Intell, PAMI-4 n6, 663–666.Google Scholar
- A. SBIHI and J. G. POSTAIRE (1995): Mode Extraction by Multivalue Morphology for Cluster Analysis. In W.Gaul and D.Pfeifer (eds) : From DATA to Knowledge: Theortical and Practical of aspect of Classification. Springer, Berlin, 212–221.Google Scholar