Fuzzy clustering for colour reduction in images

Gerald Schaefer*, Huiyu Zhou

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

The aim of colour quantisation is to reduce the number of distinct colour in images while preserving a high colour fidelity as compared to the original images. The choice of a good colour palette is crucial as it directly determines the quality of the resulting image. Colour quantisation can also be seen as a clustering problem where the task is to identify those clusters that best represent the colours in an image. In this paper we investigate the performance of various fuzzy c-means clustering algorithms for colour quantisation of images. In particular, we use conventional fuzzy c-means as well as some more efficient variants thereof, namely fast fuzzy c-means with random sampling, fast generalised fuzzy c-means, and a recently introduced anisotropic mean shift based fuzzy c-means algorithm. Experimental results show that fuzzy c-means performs significantly better than other, purpose built colour quantisation algorithms, and also confirm that the fast fuzzy clustering algorithms provide similar quantisation results to the full conventional fuzzy c-means approach.

Original languageEnglish
Pages (from-to)17-25
Number of pages9
JournalTelecommunication Systems
Volume40
Issue number1-2
Early online date11 Oct 2008
DOIs
Publication statusPublished - Feb 2009

Fingerprint

Fuzzy clustering
Color
Clustering algorithms
Sampling

Keywords

  • Colour palette
  • Colour quantisation
  • Fuzzy c-means
  • Fuzzy clustering

Cite this

Schaefer, Gerald ; Zhou, Huiyu. / Fuzzy clustering for colour reduction in images. In: Telecommunication Systems. 2009 ; Vol. 40, No. 1-2. pp. 17-25.
@article{1b4da10923b54a20b74ef15e21c9b682,
title = "Fuzzy clustering for colour reduction in images",
abstract = "The aim of colour quantisation is to reduce the number of distinct colour in images while preserving a high colour fidelity as compared to the original images. The choice of a good colour palette is crucial as it directly determines the quality of the resulting image. Colour quantisation can also be seen as a clustering problem where the task is to identify those clusters that best represent the colours in an image. In this paper we investigate the performance of various fuzzy c-means clustering algorithms for colour quantisation of images. In particular, we use conventional fuzzy c-means as well as some more efficient variants thereof, namely fast fuzzy c-means with random sampling, fast generalised fuzzy c-means, and a recently introduced anisotropic mean shift based fuzzy c-means algorithm. Experimental results show that fuzzy c-means performs significantly better than other, purpose built colour quantisation algorithms, and also confirm that the fast fuzzy clustering algorithms provide similar quantisation results to the full conventional fuzzy c-means approach.",
keywords = "Colour palette, Colour quantisation, Fuzzy c-means, Fuzzy clustering",
author = "Gerald Schaefer and Huiyu Zhou",
year = "2009",
month = "2",
doi = "10.1007/s11235-008-9143-8",
language = "English",
volume = "40",
pages = "17--25",
journal = "Telecommunication Systems",
issn = "1018-4864",
publisher = "Springer",
number = "1-2",

}

Fuzzy clustering for colour reduction in images. / Schaefer, Gerald; Zhou, Huiyu.

In: Telecommunication Systems, Vol. 40, No. 1-2, 02.2009, p. 17-25.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fuzzy clustering for colour reduction in images

AU - Schaefer, Gerald

AU - Zhou, Huiyu

PY - 2009/2

Y1 - 2009/2

N2 - The aim of colour quantisation is to reduce the number of distinct colour in images while preserving a high colour fidelity as compared to the original images. The choice of a good colour palette is crucial as it directly determines the quality of the resulting image. Colour quantisation can also be seen as a clustering problem where the task is to identify those clusters that best represent the colours in an image. In this paper we investigate the performance of various fuzzy c-means clustering algorithms for colour quantisation of images. In particular, we use conventional fuzzy c-means as well as some more efficient variants thereof, namely fast fuzzy c-means with random sampling, fast generalised fuzzy c-means, and a recently introduced anisotropic mean shift based fuzzy c-means algorithm. Experimental results show that fuzzy c-means performs significantly better than other, purpose built colour quantisation algorithms, and also confirm that the fast fuzzy clustering algorithms provide similar quantisation results to the full conventional fuzzy c-means approach.

AB - The aim of colour quantisation is to reduce the number of distinct colour in images while preserving a high colour fidelity as compared to the original images. The choice of a good colour palette is crucial as it directly determines the quality of the resulting image. Colour quantisation can also be seen as a clustering problem where the task is to identify those clusters that best represent the colours in an image. In this paper we investigate the performance of various fuzzy c-means clustering algorithms for colour quantisation of images. In particular, we use conventional fuzzy c-means as well as some more efficient variants thereof, namely fast fuzzy c-means with random sampling, fast generalised fuzzy c-means, and a recently introduced anisotropic mean shift based fuzzy c-means algorithm. Experimental results show that fuzzy c-means performs significantly better than other, purpose built colour quantisation algorithms, and also confirm that the fast fuzzy clustering algorithms provide similar quantisation results to the full conventional fuzzy c-means approach.

KW - Colour palette

KW - Colour quantisation

KW - Fuzzy c-means

KW - Fuzzy clustering

UR - http://link.springer.com/article/10.1007%2Fs11235-008-9143-8

UR - http://www.scopus.com/inward/record.url?scp=60949104695&partnerID=8YFLogxK

U2 - 10.1007/s11235-008-9143-8

DO - 10.1007/s11235-008-9143-8

M3 - Article

AN - SCOPUS:60949104695

VL - 40

SP - 17

EP - 25

JO - Telecommunication Systems

JF - Telecommunication Systems

SN - 1018-4864

IS - 1-2

ER -