[argyllcms] Re: Implement cal into profile

  • From: Graeme Gill <graeme@xxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Mon, 15 Oct 2012 22:43:45 +1100

Dimitrije Zivkovic wrote:

> QUESTION
> Can I do at once somehow?
> 
> 1. print single chanel patches and patches for profile (two separated charts)
> 2. make profile (standard)
> 3. make cal file
> 4. apply cal

Hi,

Maybe you can, if you go to enough trouble, but the result may be disappointing,
because you do not get the benefits of a better behaved device during the
profiling stage.

A sketch of how you would do it would be to print both a calibration and 
profiling
chart, create the calibration curves, and then (somehow) apply the inverse of
the calibration curves to the device values in the .ti3 file, before
creating the profile. You'd have to throw away any points that clip during the
transformation (and this may leave the edges of the gamut poorly defined).

Argyll doesn't currently have a tool to apply the inverse of a set of
calibration curves to textural data though. cctiff will do this for
a raster file, but there is no convenient way to convert from
text values to & from a raster file. It's conceivable that this
could be done using a spreadsheet.

Graeme Gill.

Other related posts: