DMCA
Mean shift, mode seeking, and clustering (1995)
| Venue: | IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE |
| Citations: | 624 - 0 self |
BibTeX
@ARTICLE{Cheng95meanshift,,
author = {Yizong Cheng},
title = {Mean shift, mode seeking, and clustering},
journal = {IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE},
year = {1995}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Mean shift, a simple iterative procedure that shifts each data point to the average of data points in its neighborhood, is generalized and analyzed in this paper. This generalization makes some k-means like clustering algorithms its special cases. It is shown that mean shift is a mode-seeking process on a surface constructed with a “shadow ” kernel. For Gaussian kernels, mean shift is a gradient mapping. Convergence is studied for mean shift iterations. Cluster analysis is treated as a deterministic problem of finding a fixed point of mean shift that characterizes the data. Applications in clustering and Hough transform are demon-trated. Mean shift is also considered as an evolutionary strategy that performs multistart global optimization.
Keyphrases
mean shift data point hough transform cluster analysis k-means clustering special case shadow kernel abstract-mean shift mode-seeking process deterministic problem multistart global optimization mean shift iteration gaussian kernel gradient descent index terms-mean shift global optimiza-tion simple iterative procedure evolutionary strategy gradient mapping fixed point
