[argyllcms] Profile Evaluation
- From: "Roger Breton" <graxx@xxxxxxxxxxxx>
- To: <argyllcms@xxxxxxxxxxxxx>
- Date: Sun, 12 Jun 2016 14:32:48 -0400
The colprof documentation states the following :
the maximum and average fit error of the input points to the resulting
profile will be reported.
This is a good guide as to whether things have gone smoothly in creating a
Depending on the type of device, and the consistency of the readings,
average errors of 5 or less, and maximum errors of 15 or less would normally
If errors are grossly higher than this, then this is an indication that
something is seriously wrong with the device measurement, or profile
This is my results :
peak err = 22.144786, avg err = 0.935990, RMS = 1.937815
I got Average error < 1 so I'm on the good side as far as the average is
My Max error (or "peak" error, I suppose?), though, was "22+" which is a
little over the "typical 15", suggests cause for concern.
I am still trying to dig behind this analysis.
First, are the reported units in "DeltaEs" of some kind? (Have to study the
concept of RMS which, according to some videos, comes from the world of
Second, how is the difference arrived at? Is colprof doing some kind of Lab
RGB followed by a RGB > Lab and comparing the Starting Lab values to the
Final Lab values?
I don't see how that could be done differently.
How would it possible for me to analyze the error distribution over the 200
patches used to create the profile?
Do I take the 200 RGB device values from the ti1 file and convert them to
Lab first, and then convert them back a second time from Lab to, what, RGB?
And then to Lab again?
I am somewhat lost as to the methodology... but I will continue my research
I know it's a complicated question...
Argyll may already have the analytical tool to answer this question?
Best / Roger
Other related posts: