[argyllcms] Re: Gamut mapping coefficients. How to vary?
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Wed, 20 Mar 2013 22:12:55 +1100
Nikolay Pokhilchenko wrote:
> Graeme, can You confirm that the coefficients requested above defined in
> maptest.c from line 166 in current 1.5.1 sources? I've didn't find another
> place with mapping parameters and declared coefficients.
maptest isn't used by the main utilities - it's just test code.
> and have recompiled ArgyllCMS. But with no success: freshly compiled colprof
> says for perceptual:
>
> Gamut Perceptual mapping weighting factor 1.000000
> Gamut Saturation mapping weighting factor 0.000000
Hi, there's two levels to this. At the lower level there are the parameters
in gamut/gammap.c. They are divided into perceptual and saturation settings,
and control the gamut mapping guide vectors. They can be set
per color sector - see gamut/nearsmth.h.
At a higher level the intent choice is mapped onto the
lower level machinery. See xicc/xicc.c, and the function
xicc_enum_gmapintent(). That selects colorspace, neutral axis
treatment, whether gamut mapping is used, and how
the perceptual and saturation pre-sets are used.
Looking at the VRML diagnostic output of the mapping vectors
(ie. colprof/colloink -P) is the most informative bit, as
well as checking what RGBsurfall.tif looks like. I do have
a set of challenging test images I use for sanity checking.
There are a whole lot of other diagnostic #defines at the
top of gammap.c
HTH
Graeme Gill.
Other related posts:
