Mode Extraction by Multivalue Morphology for Cluster Analysis
The new statistical approach to unsupervised pattern classification, developed in this paper, consists to extending the multivalue morphological concepts to multidimensional functions in order to detect the modes of the underlying probability density function, particularly when no a priori information is available as to the number of clusters and their distribution.
Unable to display preview. Download preview PDF.
- DEVIJVER, P.A. et KITTLER, J. (1982) Pattern recognition: A statistical approach. Prentice-Hall, Englewood Cliffs, 448Google Scholar
- GITMAN, I. (1973) An algorithm for nonsupervised pattern classification. IEEE Trans. Syst. Man. Cybern., SCM-3, 66–74Google Scholar
- VASSEUR, C.P.A. et POSTAIRE, J.G. (1980) A convexity testing method for cluster analysis. IEEE Trans. Syst. Man. Cyber., SMC-10, 3, 145–179Google Scholar
- VASSEUR, C.P.A. et POSTAIRE, J.G. (1981) An approximate solution to normal mixture identification with application to unsupervised pattern classification. IEEE Trans. Pattern Anal. Machine Intell., PAMI-3, 2,163–179Google Scholar
- MATHERON, G. (1975) Random sets and integral geometry. Wiley, New YorkGoogle Scholar
- SERRA, J. (1982) Image analysis and mathematical morphology. Academic Press, New YorkGoogle Scholar