[argyllcms] OT: some metrological questions

  • From: Klaus Karcher <lists@xxxxxxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Thu, 24 Jul 2008 17:21:21 +0200

Dear metrologists, statisticians and color experts,

(I know that you are here ;-)

I have some questions relating to assessment of color measurement data. I'm not very used to metrological/statistical terms and concepts, therefore please excuse my clumsy diction.

When I have spectral or tristimulus measurement data with repeated patches (e.g. 5x white, 3x red, 2x green, 10x blue, ...): what is the best method to average them?

And how can I use the redundant measurements to quantify the "quality" (repeatability, precision) of single and repeated measurements?

The euclidean distance of a point to the centroid (barycenter) seems to be a obvious approach to calculate its weight. But how do I get from this distance to the weight? And what is the right space to work in?

I guess that working with radiometric quantities (e.g. spectral reflectance) means to be closer to the sources of error and working with colorimeric quantities (CIELAB) means being closer to their visual significance.

- Lab is perceptually uniform and has has orthonormal bases, but there is no linear relation to radiometric quantities.

- XYZ is linearly related to radiometry, but neither orthonormal nor perceptually uniform, therefore euclidean distances in XYZ don't make much sense -- neither from a radiometric nor from a perceptual point of view.

- Working in a space with orthonormal color matching functions [1,2] seems to be benefitical: "The components [...] will tend to be less correlated and therefore more helpful in the context of measurement and quality control." [3]

- Working directly with the spectral data is probably the most appropriate (but expensive)-: method, but I have no idea e.g how take the correlations between the bands into account.

Thanks in advance for any help,
Klaus



[1] Jozef B. Cohen: Visual Color and Color Mixture: The Fundamental Color Space, ISBN 0-252-02549-0 <http://www.press.uillinois.edu/books/catalog/73gxb6tw9780252025495.html>

[2] <http://www.jimworthey.com/index.html>

[3] <http://www.jimworthey.com/vectorial.pdf>

Other related posts: