Colour in Context
Research group
Computer Vision Center

Natural illumination, shadows, and primate colour vision

Tom Troscianko, C. Alejandro Parraga, P. George Lovell, D.J. Tolhurst, R. Baddeley, U. Leonards
Perception, European Conference on Visual Perception (ECVP´04), Volume 33, page 45b - 2004
IF: 1.271. area: PSYCHOLOGY. Quartile: 3.
Download the publication : p161.pdf [26Ko]  
Previous research has established that the red - green opponent system in primate colour vision is optimised for encoding the spatio-chromatic properties of scenes containing red or yellow fruit/leaves on a background of green leaves. The efficient detection of such items is based on the fact that the red - green system removes shadows. This shadow-removing process operates even when the shadow has a blue or green tint from scattered light. In addition, information from the red - green system has been shown to be stable against changes in natural illumination. The invariance of the red - green system to shadows and changes in natural illumination, and its orthogonality to the yellow - blue system, imply that the yellow - blue system should be particularly efficient at encoding the difference between light directly reflected from the ground, and light which has undergone Rayleigh scattering. We present data obtained in Kibale Forest, Uganda, and in UK locations, which support the idea that the red - green system removes shadows, and the yellow - blue system responds particularly to the difference between scattered and direct light. The data are based both on a computational analysis and on psychophysical measures of scene segmentation over distances of several kilometres, in which Rayleigh scattering has a strong effect on cone responses.

Images and movies

 

BibTex references

@InProceedings\{TPL2004,
  author       = "Tom Troscianko and C. Alejandro Parraga and P. George Lovell and D.J. Tolhurst and R. Baddeley and U. Leonards",
  title        = "Natural illumination, shadows, and primate colour vision",
  booktitle    = "Perception, European Conference on Visual Perception (ECVP\´04)",
  volume       = "33",
  pages        = "45b",
  year         = "2004",
  abstract     = "Previous research has established that the red - green opponent system in primate colour vision is optimised for encoding the spatio-chromatic properties of scenes containing red or yellow fruit/leaves on a background of green leaves. The efficient detection of such items is based on the fact that the red - green system removes shadows. This shadow-removing process operates even when the shadow has a blue or green tint from scattered light. In addition, information from the red - green system has been shown to be stable against changes in natural illumination. The invariance of the red - green system to shadows and changes in natural illumination, and its orthogonality to the yellow - blue system, imply that the yellow - blue system should be particularly efficient at encoding the difference between light directly reflected from the ground, and light which has undergone Rayleigh scattering. We present data obtained in Kibale Forest, Uganda, and in UK locations, which support the idea that the red - green system removes shadows, and the yellow - blue system responds particularly to the difference between scattered and direct light. The data are based both on a computational analysis and on psychophysical measures of scene segmentation over distances of several kilometres, in which Rayleigh scattering has a strong effect on cone responses.",
  ifactor      = "1.271",
  quartile     = "3",
  area         = "PSYCHOLOGY",
  url          = "http://cat.cvc.uab.es/Public/Publications/2004/TPL2004"
}

Other publications in the database

 © 2008 Colour in context Group | Computer Vision Center. All rights reserved | Contact webmaster |  Last updated: Monday 11 May 2009     eXTReMe Tracker