Improved mean shift algorithm with heterogeneous node weights

Ji Won Yoon, Simon P. Wilson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

The conventional mean shift algorithm has been known to be sensitive to selecting a bandwidth. We present a robust mean shift algorithm with heterogeneous node weights that come from a geometric structure of a given data set. Before running MS procedure, we reconstruct un-normalized weights (a rough surface of data points) from the Delaunay Triangulation. The un-normalized weights help MS to avoid the problem of failing of misled mean shift vectors. As a result, we can obtain a more robust clustering result compared to the conventional mean shift algorithm. We also propose an alternative way to assign weights for large size datasets and noisy datasets.

Original languageEnglish
Title of host publicationProceedings - International Conference on Pattern Recognition
Pages4222-4225
Number of pages4
DOIs
Publication statusPublished - 2010 Nov 18
Externally publishedYes
Event2010 20th International Conference on Pattern Recognition, ICPR 2010 - Istanbul, Turkey
Duration: 2010 Aug 232010 Aug 26

Other

Other2010 20th International Conference on Pattern Recognition, ICPR 2010
CountryTurkey
CityIstanbul
Period10/8/2310/8/26

Fingerprint

Triangulation
Bandwidth

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Yoon, J. W., & Wilson, S. P. (2010). Improved mean shift algorithm with heterogeneous node weights. In Proceedings - International Conference on Pattern Recognition (pp. 4222-4225). [5597740] https://doi.org/10.1109/ICPR.2010.1026

Improved mean shift algorithm with heterogeneous node weights. / Yoon, Ji Won; Wilson, Simon P.

Proceedings - International Conference on Pattern Recognition. 2010. p. 4222-4225 5597740.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yoon, JW & Wilson, SP 2010, Improved mean shift algorithm with heterogeneous node weights. in Proceedings - International Conference on Pattern Recognition., 5597740, pp. 4222-4225, 2010 20th International Conference on Pattern Recognition, ICPR 2010, Istanbul, Turkey, 10/8/23. https://doi.org/10.1109/ICPR.2010.1026
Yoon JW, Wilson SP. Improved mean shift algorithm with heterogeneous node weights. In Proceedings - International Conference on Pattern Recognition. 2010. p. 4222-4225. 5597740 https://doi.org/10.1109/ICPR.2010.1026
Yoon, Ji Won ; Wilson, Simon P. / Improved mean shift algorithm with heterogeneous node weights. Proceedings - International Conference on Pattern Recognition. 2010. pp. 4222-4225
@inproceedings{8227b534c3c94334b7b4bae37d4fcd67,
title = "Improved mean shift algorithm with heterogeneous node weights",
abstract = "The conventional mean shift algorithm has been known to be sensitive to selecting a bandwidth. We present a robust mean shift algorithm with heterogeneous node weights that come from a geometric structure of a given data set. Before running MS procedure, we reconstruct un-normalized weights (a rough surface of data points) from the Delaunay Triangulation. The un-normalized weights help MS to avoid the problem of failing of misled mean shift vectors. As a result, we can obtain a more robust clustering result compared to the conventional mean shift algorithm. We also propose an alternative way to assign weights for large size datasets and noisy datasets.",
author = "Yoon, {Ji Won} and Wilson, {Simon P.}",
year = "2010",
month = "11",
day = "18",
doi = "10.1109/ICPR.2010.1026",
language = "English",
isbn = "9780769541099",
pages = "4222--4225",
booktitle = "Proceedings - International Conference on Pattern Recognition",

}

TY - GEN

T1 - Improved mean shift algorithm with heterogeneous node weights

AU - Yoon, Ji Won

AU - Wilson, Simon P.

PY - 2010/11/18

Y1 - 2010/11/18

N2 - The conventional mean shift algorithm has been known to be sensitive to selecting a bandwidth. We present a robust mean shift algorithm with heterogeneous node weights that come from a geometric structure of a given data set. Before running MS procedure, we reconstruct un-normalized weights (a rough surface of data points) from the Delaunay Triangulation. The un-normalized weights help MS to avoid the problem of failing of misled mean shift vectors. As a result, we can obtain a more robust clustering result compared to the conventional mean shift algorithm. We also propose an alternative way to assign weights for large size datasets and noisy datasets.

AB - The conventional mean shift algorithm has been known to be sensitive to selecting a bandwidth. We present a robust mean shift algorithm with heterogeneous node weights that come from a geometric structure of a given data set. Before running MS procedure, we reconstruct un-normalized weights (a rough surface of data points) from the Delaunay Triangulation. The un-normalized weights help MS to avoid the problem of failing of misled mean shift vectors. As a result, we can obtain a more robust clustering result compared to the conventional mean shift algorithm. We also propose an alternative way to assign weights for large size datasets and noisy datasets.

UR - http://www.scopus.com/inward/record.url?scp=78149480025&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78149480025&partnerID=8YFLogxK

U2 - 10.1109/ICPR.2010.1026

DO - 10.1109/ICPR.2010.1026

M3 - Conference contribution

AN - SCOPUS:78149480025

SN - 9780769541099

SP - 4222

EP - 4225

BT - Proceedings - International Conference on Pattern Recognition

ER -