Estimating the scene illumination chromaticity by using a neural network

Vlad C. Cardei, Brian Funt, Jacobus J Barnard

Research output: Contribution to journalArticle

132 Citations (Scopus)

Abstract

A neural network can learn color constancy, defined here as the ability to estimate the chromaticity of a scene's overall illumination. We describe a multilayer neural network that is able to recover the illumination chromaticity given only an image of the scene. The network is previously trained by being presented with a set of images of scenes and the chromaticities of the corresponding scene illuminants. Experiments with real images show that the network performs better than previous color constancy methods. In particular, the performance is better for images with a relatively small number of distinct colors. The method has application to machine vision problems such as object recognition, where illumination-independent color descriptors are required, and in digital photography, where uncontrolled scene illumination can create an unwanted color cast in a photograph.

Original languageEnglish (US)
Pages (from-to)2374-2386
Number of pages13
JournalJournal of the Optical Society of America A: Optics and Image Science, and Vision
Volume19
Issue number12
StatePublished - 2002

Fingerprint

Lighting
Color
Neural networks
Photography
Object recognition
Multilayer neural networks
Computer vision
Experiments

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Computer Vision and Pattern Recognition

Cite this

Estimating the scene illumination chromaticity by using a neural network. / Cardei, Vlad C.; Funt, Brian; Barnard, Jacobus J.

In: Journal of the Optical Society of America A: Optics and Image Science, and Vision, Vol. 19, No. 12, 2002, p. 2374-2386.

Research output: Contribution to journalArticle

@article{b00929234f8f4fec890ecbc1a27b14f4,
title = "Estimating the scene illumination chromaticity by using a neural network",
abstract = "A neural network can learn color constancy, defined here as the ability to estimate the chromaticity of a scene's overall illumination. We describe a multilayer neural network that is able to recover the illumination chromaticity given only an image of the scene. The network is previously trained by being presented with a set of images of scenes and the chromaticities of the corresponding scene illuminants. Experiments with real images show that the network performs better than previous color constancy methods. In particular, the performance is better for images with a relatively small number of distinct colors. The method has application to machine vision problems such as object recognition, where illumination-independent color descriptors are required, and in digital photography, where uncontrolled scene illumination can create an unwanted color cast in a photograph.",
author = "Cardei, {Vlad C.} and Brian Funt and Barnard, {Jacobus J}",
year = "2002",
language = "English (US)",
volume = "19",
pages = "2374--2386",
journal = "Journal of the Optical Society of America A: Optics and Image Science, and Vision",
issn = "1084-7529",
publisher = "The Optical Society",
number = "12",

}

TY - JOUR

T1 - Estimating the scene illumination chromaticity by using a neural network

AU - Cardei, Vlad C.

AU - Funt, Brian

AU - Barnard, Jacobus J

PY - 2002

Y1 - 2002

N2 - A neural network can learn color constancy, defined here as the ability to estimate the chromaticity of a scene's overall illumination. We describe a multilayer neural network that is able to recover the illumination chromaticity given only an image of the scene. The network is previously trained by being presented with a set of images of scenes and the chromaticities of the corresponding scene illuminants. Experiments with real images show that the network performs better than previous color constancy methods. In particular, the performance is better for images with a relatively small number of distinct colors. The method has application to machine vision problems such as object recognition, where illumination-independent color descriptors are required, and in digital photography, where uncontrolled scene illumination can create an unwanted color cast in a photograph.

AB - A neural network can learn color constancy, defined here as the ability to estimate the chromaticity of a scene's overall illumination. We describe a multilayer neural network that is able to recover the illumination chromaticity given only an image of the scene. The network is previously trained by being presented with a set of images of scenes and the chromaticities of the corresponding scene illuminants. Experiments with real images show that the network performs better than previous color constancy methods. In particular, the performance is better for images with a relatively small number of distinct colors. The method has application to machine vision problems such as object recognition, where illumination-independent color descriptors are required, and in digital photography, where uncontrolled scene illumination can create an unwanted color cast in a photograph.

UR - http://www.scopus.com/inward/record.url?scp=0041571878&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0041571878&partnerID=8YFLogxK

M3 - Article

VL - 19

SP - 2374

EP - 2386

JO - Journal of the Optical Society of America A: Optics and Image Science, and Vision

JF - Journal of the Optical Society of America A: Optics and Image Science, and Vision

SN - 1084-7529

IS - 12

ER -