Effective measurement selection in truncated kernel density estimator: Voronoi mean shift algorithm for truncated kernels

Ji Won Yoon, Hyoung Joo Lee, Hyoungshick Kim

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

1 Citation (Scopus)

Abstract

The Gating/Truncation technique is adapted to choose relatively significant measurements rather than all measurements to speed up mean shift algorithm which is one of the well-known clustering algorithms in the field of computer vision. The conventional mean shift algorithm can be sensitive to selecting measurements since the measurements are truncated with a Gaussian window of a fixed size. In particular when a small gating window is selected, it cannot properly cluster data points located far from major clusters and thus it generates unwanted, small clusters. We present a robust gating technique for truncated mean shift algorithm based on a geometric structure called Voronoi diagram of a given data set. Unlike conventional gating/truncation techniques our proposed truncation technique can provide nonlinear truncation windows with variable sizes constructed by using the Voronoi diagram to effectively identify outlier points in clusters. We also demonstrate the feasibility of this technique by applying it on synthetic and real-world image data sets. The experimental results show that the proposed truncation technique provides a more robust clustering result compared to the conventional truncation techniques. The proposed algorithm can be effectively applied to denoising of images by removing background noise.

Original languageEnglish
Title of host publicationProceedings of the 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011
DOIs
Publication statusPublished - 2011 May 20
Externally publishedYes
Event5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011 - Seoul, Korea, Republic of
Duration: 2011 Feb 212011 Feb 23

Other

Other5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011
CountryKorea, Republic of
CitySeoul
Period11/2/2111/2/23

Fingerprint

Clustering algorithms
Computer vision

Keywords

  • Clustering
  • Image processing
  • Mean shift
  • Truncated Gaussian kernel
  • Voronoi diagram

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems

Cite this

Yoon, J. W., Lee, H. J., & Kim, H. (2011). Effective measurement selection in truncated kernel density estimator: Voronoi mean shift algorithm for truncated kernels. In Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011 [57] https://doi.org/10.1145/1968613.1968683

Effective measurement selection in truncated kernel density estimator : Voronoi mean shift algorithm for truncated kernels. / Yoon, Ji Won; Lee, Hyoung Joo; Kim, Hyoungshick.

Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011. 2011. 57.

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

Yoon, JW, Lee, HJ & Kim, H 2011, Effective measurement selection in truncated kernel density estimator: Voronoi mean shift algorithm for truncated kernels. in Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011., 57, 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011, Seoul, Korea, Republic of, 11/2/21. https://doi.org/10.1145/1968613.1968683
Yoon JW, Lee HJ, Kim H. Effective measurement selection in truncated kernel density estimator: Voronoi mean shift algorithm for truncated kernels. In Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011. 2011. 57 https://doi.org/10.1145/1968613.1968683
Yoon, Ji Won ; Lee, Hyoung Joo ; Kim, Hyoungshick. / Effective measurement selection in truncated kernel density estimator : Voronoi mean shift algorithm for truncated kernels. Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011. 2011.
@inproceedings{64a45cc13a9441a79baa5e0188def7d1,
title = "Effective measurement selection in truncated kernel density estimator: Voronoi mean shift algorithm for truncated kernels",
abstract = "The Gating/Truncation technique is adapted to choose relatively significant measurements rather than all measurements to speed up mean shift algorithm which is one of the well-known clustering algorithms in the field of computer vision. The conventional mean shift algorithm can be sensitive to selecting measurements since the measurements are truncated with a Gaussian window of a fixed size. In particular when a small gating window is selected, it cannot properly cluster data points located far from major clusters and thus it generates unwanted, small clusters. We present a robust gating technique for truncated mean shift algorithm based on a geometric structure called Voronoi diagram of a given data set. Unlike conventional gating/truncation techniques our proposed truncation technique can provide nonlinear truncation windows with variable sizes constructed by using the Voronoi diagram to effectively identify outlier points in clusters. We also demonstrate the feasibility of this technique by applying it on synthetic and real-world image data sets. The experimental results show that the proposed truncation technique provides a more robust clustering result compared to the conventional truncation techniques. The proposed algorithm can be effectively applied to denoising of images by removing background noise.",
keywords = "Clustering, Image processing, Mean shift, Truncated Gaussian kernel, Voronoi diagram",
author = "Yoon, {Ji Won} and Lee, {Hyoung Joo} and Hyoungshick Kim",
year = "2011",
month = "5",
day = "20",
doi = "10.1145/1968613.1968683",
language = "English",
isbn = "9781450305716",
booktitle = "Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011",

}

TY - GEN

T1 - Effective measurement selection in truncated kernel density estimator

T2 - Voronoi mean shift algorithm for truncated kernels

AU - Yoon, Ji Won

AU - Lee, Hyoung Joo

AU - Kim, Hyoungshick

PY - 2011/5/20

Y1 - 2011/5/20

N2 - The Gating/Truncation technique is adapted to choose relatively significant measurements rather than all measurements to speed up mean shift algorithm which is one of the well-known clustering algorithms in the field of computer vision. The conventional mean shift algorithm can be sensitive to selecting measurements since the measurements are truncated with a Gaussian window of a fixed size. In particular when a small gating window is selected, it cannot properly cluster data points located far from major clusters and thus it generates unwanted, small clusters. We present a robust gating technique for truncated mean shift algorithm based on a geometric structure called Voronoi diagram of a given data set. Unlike conventional gating/truncation techniques our proposed truncation technique can provide nonlinear truncation windows with variable sizes constructed by using the Voronoi diagram to effectively identify outlier points in clusters. We also demonstrate the feasibility of this technique by applying it on synthetic and real-world image data sets. The experimental results show that the proposed truncation technique provides a more robust clustering result compared to the conventional truncation techniques. The proposed algorithm can be effectively applied to denoising of images by removing background noise.

AB - The Gating/Truncation technique is adapted to choose relatively significant measurements rather than all measurements to speed up mean shift algorithm which is one of the well-known clustering algorithms in the field of computer vision. The conventional mean shift algorithm can be sensitive to selecting measurements since the measurements are truncated with a Gaussian window of a fixed size. In particular when a small gating window is selected, it cannot properly cluster data points located far from major clusters and thus it generates unwanted, small clusters. We present a robust gating technique for truncated mean shift algorithm based on a geometric structure called Voronoi diagram of a given data set. Unlike conventional gating/truncation techniques our proposed truncation technique can provide nonlinear truncation windows with variable sizes constructed by using the Voronoi diagram to effectively identify outlier points in clusters. We also demonstrate the feasibility of this technique by applying it on synthetic and real-world image data sets. The experimental results show that the proposed truncation technique provides a more robust clustering result compared to the conventional truncation techniques. The proposed algorithm can be effectively applied to denoising of images by removing background noise.

KW - Clustering

KW - Image processing

KW - Mean shift

KW - Truncated Gaussian kernel

KW - Voronoi diagram

UR - http://www.scopus.com/inward/record.url?scp=79955993904&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955993904&partnerID=8YFLogxK

U2 - 10.1145/1968613.1968683

DO - 10.1145/1968613.1968683

M3 - Conference contribution

AN - SCOPUS:79955993904

SN - 9781450305716

BT - Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication, ICUIMC 2011

ER -