Analyzing an image or other data to obtain a stable number of groups

作者: Satyajit Rao , James V. Mahoney

DOI:

关键词: Image (mathematics)StatisticsRunning totalSimilarity (geometry)MathematicsSegmentationSet (abstract data type)Group (mathematics)Bounded functionRange (statistics)

摘要: To group items in an array, gap data are obtained indicating gaps between items. The used to obtain threshold data, which then grouping data. could, for example, be distances a two-dimensional array or differences values at occur one-dimensional array. indicate threshold. would produce number of groups the that is stable across range thresholds, and thresholds meets criterion largeness range. can require, larger than any other set numbers groups. iteratively by applying candidate each iteration. incremented, iterations counted find meeting criterion. Or increased gaps, running sum ranges also directly finding largest difference extents obtaining within difference. Many types performed, including spatial clustering, segmentation partially bounded regions, local width, global similarity grouping.

参考文章(13)
Young-Chul Kim, Min-Ki Yeo, Dong-Yool Kim, Yong-Woo Park, Hideki Koide, Hong-Jik Lee, Document image signal processor having an adaptive threshold ,(1990)
Lop G. Ng, Teresa A. Hinkle, deceased Sacks, Thomas K. Lo, Automatic mask threshold ,(1991)
John C. Russ, The Image Processing Handbook ,(1992)
Richard C. Dubes, Anil K. Jain, Algorithms for clustering data ,(1988)
A. Witkin, Scale-space filtering: A new approach to multi-scale description international conference on acoustics, speech, and signal processing. ,vol. 9, pp. 150- 153 ,(1984) , 10.1109/ICASSP.1984.1172729
Michael A. Bickel, Adaptive fast fuzzy clustering system ,(1991)