[argyllcms] Re: How to create deltaE plot?
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Sat, 11 Jun 2016 10:26:34 +1000
But where do I find the measured x,y and where the ones corrected with
by themselves, device color profiles don't "correct" anything. They
the color behaviour of a device. To make one device emulate another, or to
color from one device space to another requires linking two device profiles to
a workflow. Typically this is done by the system or application CMM, or can be
explicitly by (say) a tool like collink.
So you can check the accuracy of the colorimetric tables of a profile using
a tool like profcheck, or you can check the accuracy of a workflow using
a tool such as colverify, although the latter requires understanding
the workflow sufficiently to figure out how to create a file that
contains a set of color reference points that represent the desired
To truly check a profile, one would measure the device a second
time (after profiling) with an even larger set of differently
located test points, and use that as the input to colverify.
Checking the accuracy of the reverse tables (i.e. B2A direction)
is somewhat trickier, since one expects errors when out of gamut
colors are transformed. Observing the characteristics of intents
other than colormetric is also less straightforward, since
perceptual and saturation intents have "creative" or "preferential"
transformations applied to them.
Other related posts: