Metamerism (Metameric Pair)
By David Crowther
Definition - The phenomenon where two colours appear to match under on light source, yet do not match under a different light source. Two such colours are called a metameric pair.
Any discussion on colour management will usually bring up some sort of talk or explanation on metamerism.
Producers of photographic and fine art prints often encounter a situation whereby viewing a print illuminated by one light source produces acceptable colour. By simply moving that print into a different location with a different type of lighting, a severe colour shift may appear. That is the downside of metamerism, but the upside is that for colour management to work, metamerism is what makes colour reproduction possible. Metamerism lets us display vivid yellows or flesh tones on a computer display without having to use out-and-out yellow or flesh tone colour phosphors.
Without metamerism we would have to reproduce colours by duplicating the exact spectral content of the original colour stimulus.
Only by using spectral data do we have the ability to predict the effects of different light sources on an object’s appearance. Different light sources have their own compositions of wavelengths, which in turn are affected by the object in different ways. For example, have you ever matched a pair of socks and pants under fluorescent department store lighting, then later discover that they do not match as well under your home’s incandescent lighting? This phenomenon is called metamerism.
The diagram below compares two shades of grey that are a metameric match.
Under daylight conditions, these greys appear to match quite well. However, under incandescent lighting, the first grey sample takes on a reddish cast. These changes can be demonstrated by plotting the spectral curves for the different greys and the different illuminants, then comparing where their strongest reflectance power occurs in relation to each other and to the visible spectrum wavelengths.
When our samples are illuminated by daylight, the relationship between these two colours is enhanced in the blue region (the highlighted area), where the curves are close together. Incandescent light, on the other hand, distributes more reflectance power in the red region, where the two sample curves happen to separate sharply.
So, under cooler lighting the differences between the two samples are not so apparent; but the differences are quite apparent when viewed in warmer lighting. Our eyesight can be fooled by these changes in lighting conditions. Because tristimulus data is illuminant-dependent, these formats cannot demonstrate the effects of these changes. Only spectral data can clearly detect these characteristics.
As mentioned at the outset metamerism is a common issue with many output devices. We may have what is considered to be an excellent output ICC profile that shows beautiful neutral greys in one lighting condition and then the same greys look completely colour casted when simply moved into a different location with a different illuminant. This effect really cannot be removed, but rather compensated for. Ideally, all our prints would be viewed a stable standard light source like D50. Most ICC profiles assume a viewing condition of D50. However, we know that in reality there are many kinds of light sources, and often mixtures of lighting conditions. One method of compensation involves measuring the final light under which a print will be viewed, thereby building that into the output profile. This provides predictable colour appearance using colour management techniques.
Note: The information and images contained in this article are reproduced with the permission of X-Rite Inc.