Touzani, A.; Postaire, J.-G. Clustering by mode boundary detection. (English) Zbl 0658.62075 Pattern Recognit. Lett. 9, No. 1, 1-12 (1989). A new approach to clustering is developed, which is based on detecting the boundaries of the modes of the underlying probability density function. This approach calls for a generalization of the concepts of boundary to multidimensional functions. It is particularly useful when no a priori information is available as to the number of clusters and their distribution. Cited in 1 Document MSC: 62H30 Classification and discrimination; cluster analysis (statistical aspects) 68T10 Pattern recognition, speech recognition Keywords:mode detection; boundary detection; median filter; edge operator; connectivity; nonparametric estimation; nonlinear smoothing; pattern recognition; clustering; boundary to multidimensional functions PDFBibTeX XMLCite \textit{A. Touzani} and \textit{J. G. Postaire}, Pattern Recognit. Lett. 9, No. 1, 1--12 (1989; Zbl 0658.62075) Full Text: DOI