CIE Colour Systems Questions and Answers

Answer: C

The first two are correct and follow from the definition. The third is contradicted directly by the normalisation step in the transformation. White point dependence is built in from the start. Normalisation is the first step of the transform.

Answer: D

Correct answer: dividing by the white point approximates von Kries adaptation

Each of X, Y and Z is divided by the corresponding value for the white before the cube root is taken, which scales the responses to the illumination. That is a crude von Kries transform performed in XYZ rather than in a cone space, and it is why CIELAB values are comparable across illuminants at all.

Answer: A

Both are true and the reason gives the defect, and the error is largest for strongly coloured illuminants. CAT02 and its successors operate in a cone response space instead. The error grows with the colour of the illuminant.

Answer: B

Correct answer: retains a chromaticity diagram with straight line mixing

CIELUV keeps the projective relationship to XYZ, so additive mixtures still lie on straight lines and a uniform chromaticity diagram can be drawn. That suits the display and lighting industries, where colours are added. CIELAB gives up the straight lines and is preferred for surface colours, where they are of no use.

Answer: C

The first two are correct and reflect actual usage. The third is not so, and the two spaces give measurably different differences for the same pair. Neither is a simple relabelling of the other. Their difference values genuinely differ.

Answer: D

Both are true and the reason gives the reason for the division. Additive mixing matters for displays and hardly at all for dyed textiles. Application decides which space is appropriate. Displays and textiles have different needs.

Answer: A

Saturation requires a chromaticity measure independent of lightness. CIELAB provides chroma but no corresponding saturation correlate. A chromaticity measure is needed for saturation. CIELAB provides chroma but not saturation.

Answer: B

The two spaces share the lightness function and differ in their chromatic parts. Saturation is the one correlate available in only one of them. Lightness and hue angle appear in both. Only saturation is confined to one space.

Answer: C

Correct answer: the cube root of the luminance factor relative to white

\[ L^* = 116\left(\frac{Y}{Y_n}\right)^{1/3} - 16, \] with a linear segment near black to avoid the infinite slope there. The two spaces share this function exactly, so a colour has the same lightness in both and they differ only in how they treat chromaticity.

Answer: D

Both are true and the reason gives the reason, and the crossover occurs at a very low value. Numerical behaviour near black would otherwise be poor. An infinite slope at zero would be troublesome. The crossover occurs at a very low value.