A comparison of computational color constancy algorithms - Part I: Methodology and experiments with synthesized data

Jacobus J Barnard, Vlad Cardei, Brian Funt

Research output: Contribution to journalArticle

313 Citations (Scopus)

Abstract

We introduce a context for testing computational color constancy, specify our approach to the implementation of a number of the leading algorithms, and report the results of three experiments using synthesized data. Experiments using synthesized data are important because the ground truth is known, possible confounds due to camera characterization and pre-processing are absent, and various factors affecting color constancy can be efficiently investigated because they can be manipulated individual and precisely. The algorithms chosen for close study include two gray world methods, a limiting case of a version of the Retinex method, a number of variants of Forsyth's gamut-mapping method, Cardei et al.'s neural net method, and Finlayson et al.'s Color by Correlation method. We investigate the ability of these algorithms to make estimates of three different color constancy quantities: the chromaticity of the scene illuminant, the overall magnitude of that illuminant, and a corrected, illumination invariant, image. We consider algorithm performance as a function of the number of surfaces in scenes generated from reflectance spectra, the relative effect on the algorithms of added specularities, and the effect of subsequent clipping of the data. All data is available on-line at http://www.cs.sfu.ca/∼color/data, and implementations for most of the algorithms are also available (http://www.cs.sfu.ca/∼color/code).

Original languageEnglish (US)
Pages (from-to)972-984
Number of pages13
JournalIEEE Transactions on Image Processing
Volume11
Issue number9
DOIs
StatePublished - Sep 2002
Externally publishedYes

Fingerprint

Color Constancy
Color
Methodology
Experiment
Experiments
Color codes
Clipping
Neural Nets
Correlation methods
Reflectance
Preprocessing
Illumination
Limiting
Lighting
Camera
Cameras
Neural networks
Testing
Invariant
Processing

Keywords

  • Algorithm
  • Color by correlation
  • Color constancy
  • Comparison
  • Computational
  • Gamut constraint
  • Neural network

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Graphics and Computer-Aided Design
  • Software
  • Theoretical Computer Science
  • Computational Theory and Mathematics
  • Computer Vision and Pattern Recognition

Cite this

A comparison of computational color constancy algorithms - Part I : Methodology and experiments with synthesized data. / Barnard, Jacobus J; Cardei, Vlad; Funt, Brian.

In: IEEE Transactions on Image Processing, Vol. 11, No. 9, 09.2002, p. 972-984.

Research output: Contribution to journalArticle

@article{b4e3e8f6eaea4aca92798385d937457d,
title = "A comparison of computational color constancy algorithms - Part I: Methodology and experiments with synthesized data",
abstract = "We introduce a context for testing computational color constancy, specify our approach to the implementation of a number of the leading algorithms, and report the results of three experiments using synthesized data. Experiments using synthesized data are important because the ground truth is known, possible confounds due to camera characterization and pre-processing are absent, and various factors affecting color constancy can be efficiently investigated because they can be manipulated individual and precisely. The algorithms chosen for close study include two gray world methods, a limiting case of a version of the Retinex method, a number of variants of Forsyth's gamut-mapping method, Cardei et al.'s neural net method, and Finlayson et al.'s Color by Correlation method. We investigate the ability of these algorithms to make estimates of three different color constancy quantities: the chromaticity of the scene illuminant, the overall magnitude of that illuminant, and a corrected, illumination invariant, image. We consider algorithm performance as a function of the number of surfaces in scenes generated from reflectance spectra, the relative effect on the algorithms of added specularities, and the effect of subsequent clipping of the data. All data is available on-line at http://www.cs.sfu.ca/∼color/data, and implementations for most of the algorithms are also available (http://www.cs.sfu.ca/∼color/code).",
keywords = "Algorithm, Color by correlation, Color constancy, Comparison, Computational, Gamut constraint, Neural network",
author = "Barnard, {Jacobus J} and Vlad Cardei and Brian Funt",
year = "2002",
month = "9",
doi = "10.1109/TIP.2002.802531",
language = "English (US)",
volume = "11",
pages = "972--984",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "9",

}

TY - JOUR

T1 - A comparison of computational color constancy algorithms - Part I

T2 - Methodology and experiments with synthesized data

AU - Barnard, Jacobus J

AU - Cardei, Vlad

AU - Funt, Brian

PY - 2002/9

Y1 - 2002/9

N2 - We introduce a context for testing computational color constancy, specify our approach to the implementation of a number of the leading algorithms, and report the results of three experiments using synthesized data. Experiments using synthesized data are important because the ground truth is known, possible confounds due to camera characterization and pre-processing are absent, and various factors affecting color constancy can be efficiently investigated because they can be manipulated individual and precisely. The algorithms chosen for close study include two gray world methods, a limiting case of a version of the Retinex method, a number of variants of Forsyth's gamut-mapping method, Cardei et al.'s neural net method, and Finlayson et al.'s Color by Correlation method. We investigate the ability of these algorithms to make estimates of three different color constancy quantities: the chromaticity of the scene illuminant, the overall magnitude of that illuminant, and a corrected, illumination invariant, image. We consider algorithm performance as a function of the number of surfaces in scenes generated from reflectance spectra, the relative effect on the algorithms of added specularities, and the effect of subsequent clipping of the data. All data is available on-line at http://www.cs.sfu.ca/∼color/data, and implementations for most of the algorithms are also available (http://www.cs.sfu.ca/∼color/code).

AB - We introduce a context for testing computational color constancy, specify our approach to the implementation of a number of the leading algorithms, and report the results of three experiments using synthesized data. Experiments using synthesized data are important because the ground truth is known, possible confounds due to camera characterization and pre-processing are absent, and various factors affecting color constancy can be efficiently investigated because they can be manipulated individual and precisely. The algorithms chosen for close study include two gray world methods, a limiting case of a version of the Retinex method, a number of variants of Forsyth's gamut-mapping method, Cardei et al.'s neural net method, and Finlayson et al.'s Color by Correlation method. We investigate the ability of these algorithms to make estimates of three different color constancy quantities: the chromaticity of the scene illuminant, the overall magnitude of that illuminant, and a corrected, illumination invariant, image. We consider algorithm performance as a function of the number of surfaces in scenes generated from reflectance spectra, the relative effect on the algorithms of added specularities, and the effect of subsequent clipping of the data. All data is available on-line at http://www.cs.sfu.ca/∼color/data, and implementations for most of the algorithms are also available (http://www.cs.sfu.ca/∼color/code).

KW - Algorithm

KW - Color by correlation

KW - Color constancy

KW - Comparison

KW - Computational

KW - Gamut constraint

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=0036709479&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036709479&partnerID=8YFLogxK

U2 - 10.1109/TIP.2002.802531

DO - 10.1109/TIP.2002.802531

M3 - Article

C2 - 18249720

AN - SCOPUS:0036709479

VL - 11

SP - 972

EP - 984

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 9

ER -