Spectral tools - Metamerism
|
|
|
|
Two colors which have different spectrums can produce the same visual color stimuli; this property is called metamerism. This is a very powerful and helpful phenomenon. For example, your color monitor can reproduce the color of green grass using a completely different spectrum than the one measured for the same grass. However, metamerism can also play tricks on us. A color sample measured under one light source may look quite different under another light; this "inconstancy" can be evaluated using the "Color Inconstancy Index" (CII) which computes the difference between the perceived color under a given illuminant (or a user-measured light source) and a reference illuminant (D65 is used as default but you can select among 12 preset illuminants, or the user-measured light source). The CII is computed using the CIECAT02 Chromatic Adaptation Transform (CAT) and standard color-difference formulas, such as CIE94 and CIEDE2000. Similarly, two color samples matching under a given light may not match under different lighting conditions; the color difference computed under the non-matching illuminant is then called the "Special Metamerism Index" (SMI). Finally, when evaluating a pair of samples under one light condition, even if they do not match, we might want to see if they have a similar relation under another light; this is the generic "Metamerism Index" (MI). In the screenshot above, we measured two printed grey patches, one made from black ink only (the Reference) and the other from a mix of colored inks (the Sample). These patches can be found on page 20 of the classic book Principles of Color Technology, revised by Roy Berns (3rd ed.). We compared them using measured ambient light (coming from a window, with a color temperature of 6020 K), and Illuminant A, typical of a yellowish tungsten lamp. The patches look very similar under the ambient light but we can see that the grey from mixed colored inks becomes reddish under Illuminant A. The CII for the patches under the ambient light is very small (0,05 and 0,16), indicating that the color would be quite similar under either D65, the selected CII reference, or the ambient light. The CII for the Reference patch under Illuminant A is also small, indicating there also that it would look the same when observed with Illuminant A or with the D65 CII reference. However, we see that the CII for the Sample patch is high for Illuminant A, corresponding to the visible red shift. The DeltaE* between the Reference and the Sample for each selected illuminant (ambient light and Illuminant A) is shown on the right; it is computed with the selected color-difference formula (CIE94 (2:2)). If we had a perfect match under the ambient light, we could use the DeltaE* value computed for Illuminant A as the SMI; this is not the case, but we can see the increased color difference under Illuminant A. The overall Metamerism Index (MI) is obtained with the L*a*b* values of the Reference and Sample under each Illuminant, for a total of four L*a*b* data sets; it is relatively high at 2,41. The overall result for these two patches, with the analysis we have done so far, is that we should restrain their use to the D60/D65 range, and not show them under tungsten lighting, if we expect to perceived them as identical. |
|
Legal Information -
Privacy Policy -
About us -
Press/Media
|
