Color Naming

In order to establish a precise colour identity for any given name, CIE hue angle (h*) works well enough as a definition of Hue. However, it ignores the other two dimensions of the sensation i.e. Chroma (or purity, vividness) and Lightness as for example in a ‘pale dull green’. It also becomes meaningless at the achromatic grey scale, and fails to denote a Brown colour (which is in fact a very dark orange).

For a comprehensive naming system I would refer you to the concept of a colour atlas withh llustrating samplesi such a those available fors the Inter Society Color Council-National Bureau of Standards (or ISCC-NBS) colour naming system; and the following sources:-

RS Berns’ book ‘Billmeyer and Saltzman's principles of color technology’: Published in 2000 by Wiley New York.

Keith Kelly’s book ‘Color: universal language and dictionary of names’ KL Kelly, DB Judd - 1976 - books.google.com,

“A method for color naming and description of color composition in images” A. Mojsilovic Proceedings. International Conference on Image Processing, Rochester, NY, USA, 22-25 Sept. 2002: ISBN: 0-7803-7622-6 Publ. IEEE

Thank you very much for your answer. Of course, you are right in your considerations.

But this project deals only with chromatic colors and color group's placement in the CIELAB a* b* Projection Circle.

In short, I try to reinvent the (color) wheel. Newton, Goethe and Munsell all had clear boundaries between their color groups in their color wheels. For example between, red, orange and yellow. I am trying to find out if it makes any sense to reinstate these clear limits in CIELAB - First and foremost, if it is possible at all.

Thanks for suggestions for literature. I actually have Billmeyer and Saltzman

‘morning Michael,

I warmly support your idea of updating the Color Wheel and your choice of (h*) hue angle and the CIELAB a* b* Projection Circle to represent it. Clearly in any representation of the Hue change continuum using physical samples, visible boundaries are a practical necessity. This in turn establishes a need to quantify them as visual differences. I suggest therefore that where the boundaries lie is an aesthetic matter rather than one to be specified in numerically precise terms. How about at ‘steps of visually equivalent size’? This is of course exactly what the CIELAB model is attempting in the first place; are we going round in circles here?

Thank you for your encouragement.

Yes, it was probably the original idea with DeltaEab, but it has been replaced by DeltaE00 which takes into account the color difference we actually see - and not just the distance between two points (DEab).

But one thing is to have a color difference value and a change in hue, another thing is to determine if this change is due to the fact that the color have changed color group - from the yellow color group to the orange color group.

However, this color circle problem has existed for more than 300 years, so there is probably no obvious solution.

But maybe a new way of thinking?