[argyllcms] Re: How is delta E calculated in colprof?

  • From: Graeme Gill <graeme@xxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Sat, 21 Jul 2012 20:08:16 +1000

Stephen T wrote:
> Here are some examples. The input data is a direct cut-and-paste from the 
> colprof output. I have
> calculated delta E with Gaurav Sharma's spreadsheet and Bruce Lindblooms web 
> calculator: 

> I must have not noticed before that the delta E reported by colprof (flag -y) 
> exactly matches
> DE76. Anyone can easily verify this with their own data and calculation of 
> delta E for just one
> patch will do. Graeme has also confirmed that colprof returns DE76 (Euclidean 
> distance).

As expected. There's no point in cluttering colprof with options to select the 
delta E type,
when profcheck already offers this functionality.

> Is DE76 used only for reporting errors or is it used in colprof for model 
> fitting as well?

In general DECIE94 is used for fitting, although it's often nowhere as simple 
as that
(ie., it's not applicable to cLUT profiles because the algorithms don't work
that way, and it's causes problems with per channel optimisation subroutines, 
etc., etc.)

> Which delta E function is most appropriate for profiling (my application is 
> graphic arts)? Note
> there's a huge difference between DE76 = 11.99 and DE00 = 2.77 for the last 
> patch in the above
> examples.

Overall though, it doesn't usually make a dramatic difference to the profile. 
the more advanced DE's reduce the weighting for saturated colors while leaving
neutral results unchanged. There are some situations in which DE CIE94 simply 
work within certain aspects of profile fitting due to numerical/behavioural 
and DE CIE2000 is completely out of the question due to its complex behaviour 
slow computation speed.

Graeme Gill.

Other related posts: