This is the fourteenth in a series of posts on color reproduction. The series starts here.
When I started this series of posts, I wanted to educate you all about the nature of the color reproduction problem, what some of the common — and not-so-common — techniques for dealing with non-Luther cameras are, what pitfalls are likely to be encountered, and how to do color accuracy testing for yourself. In this last objective, I limited myself to tools that you could find on the web, like Bruce Lindbloom’s Macbeth color checker spreadsheet.
That’s fine for a few spot checks, but some of the issues, like the effect of varying the illuminant, needs a fair amount of samples. Transferring data points by hand from Ps to a spreadsheet is too error prone and too slow, not to mention way too boring. So, to proceed further, I’m going to move to Matlab. I’ll be happy to post my code as I’ve done in the past, but I know that the ability of many of you to follow along at home will be adversely affected. Sorry.
There will be a hiatus in this series while I write some Matlab code and enjoy the holidays with my family. I will fill in with some things that are less demanding of my time.
Lynn Allan says
I don’t know if the following is relevant or not.
I use the QpCard-203 and free QpCalibration for custom camera profiles. Seems to work at least reasonably well for repro-graphics’ish accuracy. Or not?
The software does provide an “Inspect” capability that claims to show the De2k for each of the 35 patches, and overall.
Entertaining, but this non-color scientist has no way to check it’s accuracy.
Here’s a link to a screen-shot from QpCalibration: