Alliance for Lighting Information


Color Contrast

by David M. Keith

Color contrast is what we see - when we are photopically adapted. Color contrast may occur because of differences in luminance or chromaticity. Therefore, color contrast includes the luminance contrast which is usually meant when "contrast" is discussed, but color contrast extends further to recognize the chromatic difference between surfaces.

This extension of the information used for determining contrast provides better understanding of surface visibility than merely using luminance contrast, which is based on the "lumen". Because the "lumen" is a weighted summation over the visible spectrum, it is "spectrally ignorant", taking no account of the relationships between the chromatic characteristics of surfaces.

When luminance contrast is calculated using differences in "lumen-ous" quantities, the particular interaction of the surface with the source's SPD is lost, leading to errors in the evaluation of contrast. Furthermore any differences in chromaticity are entirely ignored, and in some cases this may be a significant aspect of what makes something visible. For example, if two surfaces have the same luminous reflectance - the same ratio of lumens-off to lumens-on - then under"equal" illuminance they should have no luminance contrast - they should not be "distinguishable" - but the reality is that they may appear quite different. If the surfaces are equal value but different colors and/or if the sources have the same "lumens" but very different SPD's, then the surfaces will have chromatic contrast but not luminance contrast. Using color contrast is the appropriate approach to evaluating such situations.

The IESNA-recommended technique for evaluating color contrast is using a uniform color space, like the CIE L*a*b*, and evaluating the overall distance within the color space between two SPD's, created by reflecting the same source SPD off two different surfaces. Research into this approach and the results from evaluating roadway signage using color contrast techniques by J. Knox and D. Keith led to a presentation and corresponding paper.

Results from this work indicate that significant differences in the calculated contrast will occur for chromatic surfaces under different sources, particularly HID. Although this work addresses signage, the specific results may not be applicable because the retro-reflection of such sign materials is so dominant. Even so, the techniques demonstrated and trends illustrated in this work do provide useful information. As an example, from such work it is clear that the yellow striping on roadways will have greater contrast against concrete or asphalt when seen under HPS than when seen under MH or high CCT fluorescent, and about the same as under incandescent sources, like most headlights. This understanding arises in part from recognition of color contrast as more indicative of overall contrast, and in part from the use of spectrally informed procedures that are necessary to properly evaluate color contrast.


This site is provided as a public service by resodance publishing co. and no guarantee is implied or provided.


This page is served & maintained by &
last changed on 28 Sep 04 by