A clustering technique based on the distance transform

Authors:
V. Starovoitov
Affiliations:
-
Venue:
Pattern Recognition Letters
Year:
1996

Citing 4
Cited 1

How many clusters are best?—an experiment

Pattern Recognition
Cluster detection in background noise

Pattern Recognition
Cluster Analysis by Binary Morphology

IEEE Transactions on Pattern Analysis and Machine Intelligence
Detection and Separation of Ring-Shaped Clusters Using Fuzzy Clustering

IEEE Transactions on Pattern Analysis and Machine Intelligence

2D Euclidean distance transform algorithms: A comparative survey

ACM Computing Surveys (CSUR)

Quantified Score

Hi-index	0.10

Visualization

Abstract

We present a new variant of a geometric approach to unsupervised clustering. It is based on the digital distance and cost function transforms. We map the given set of real continuous data onto an n-dimensional binary image, where black pixels correspond to the observations. A way of such discretization is suggested. Domains with some concentration of black pixels are extracted as cluster cores. Clusters are detected by the mentioned transforms.