DBCAMM: A novel density based clustering algorithm via using the Mahalanobis metric

In this paper we propose a new density based clustering algorithm via using the Mahalanobis metric. This is motivated by the current state-of-the-art density clustering algorithm DBSCAN and some fuzzy clustering algorithms. There are two novelties for the proposed algorithm: One is to adopt the Maha...

Full description

Bibliographic Details
Main Authors: Ren, Yan, Xiaodong, Liu, Liu, Wan-Quan
Format: Journal Article
Published: Elsevier 2012
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/33348
_version_ 1848753920916062208
author Ren, Yan
Xiaodong, Liu
Liu, Wan-Quan
author_facet Ren, Yan
Xiaodong, Liu
Liu, Wan-Quan
author_sort Ren, Yan
building Curtin Institutional Repository
collection Online Access
description In this paper we propose a new density based clustering algorithm via using the Mahalanobis metric. This is motivated by the current state-of-the-art density clustering algorithm DBSCAN and some fuzzy clustering algorithms. There are two novelties for the proposed algorithm: One is to adopt the Mahalanobis metric as distance measurement instead of the Euclidean distance in DBSCAN and the other is its effective merging approach for leaders and followers defined in this paper. This Mahalanobis metric is closely associated with dataset distribution. In order to overcome the unique density issue in DBSCAN, we propose an approach to merge the sub-clusters by using the local sub-cluster density information. Eventually we show how to automatically and efficiently extract not only ‘traditional’ clustering information, such as representative points, but also the intrinsic clustering structure. Extensive experiments on some synthetic datasets show the validity of the proposed algorithm. Further the segmentation results on some typical images by using the proposed algorithm and DBSCAN are presented in this paper and they are shown that the proposed algorithm can produce much better visual results in image segmentation.
first_indexed 2025-11-14T08:32:11Z
format Journal Article
id curtin-20.500.11937-33348
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T08:32:11Z
publishDate 2012
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-333482017-09-13T15:30:51Z DBCAMM: A novel density based clustering algorithm via using the Mahalanobis metric Ren, Yan Xiaodong, Liu Liu, Wan-Quan Mahalanobis distance Followers Image segmentation Clustering Leaders In this paper we propose a new density based clustering algorithm via using the Mahalanobis metric. This is motivated by the current state-of-the-art density clustering algorithm DBSCAN and some fuzzy clustering algorithms. There are two novelties for the proposed algorithm: One is to adopt the Mahalanobis metric as distance measurement instead of the Euclidean distance in DBSCAN and the other is its effective merging approach for leaders and followers defined in this paper. This Mahalanobis metric is closely associated with dataset distribution. In order to overcome the unique density issue in DBSCAN, we propose an approach to merge the sub-clusters by using the local sub-cluster density information. Eventually we show how to automatically and efficiently extract not only ‘traditional’ clustering information, such as representative points, but also the intrinsic clustering structure. Extensive experiments on some synthetic datasets show the validity of the proposed algorithm. Further the segmentation results on some typical images by using the proposed algorithm and DBSCAN are presented in this paper and they are shown that the proposed algorithm can produce much better visual results in image segmentation. 2012 Journal Article http://hdl.handle.net/20.500.11937/33348 10.1016/j.asoc.2011.12.015 Elsevier restricted
spellingShingle Mahalanobis distance
Followers
Image segmentation
Clustering
Leaders
Ren, Yan
Xiaodong, Liu
Liu, Wan-Quan
DBCAMM: A novel density based clustering algorithm via using the Mahalanobis metric
title DBCAMM: A novel density based clustering algorithm via using the Mahalanobis metric
title_full DBCAMM: A novel density based clustering algorithm via using the Mahalanobis metric
title_fullStr DBCAMM: A novel density based clustering algorithm via using the Mahalanobis metric
title_full_unstemmed DBCAMM: A novel density based clustering algorithm via using the Mahalanobis metric
title_short DBCAMM: A novel density based clustering algorithm via using the Mahalanobis metric
title_sort dbcamm: a novel density based clustering algorithm via using the mahalanobis metric
topic Mahalanobis distance
Followers
Image segmentation
Clustering
Leaders
url http://hdl.handle.net/20.500.11937/33348