[argyllcms] Re: noise and gamut, input profiles with B2A table

Hi,

Graeme Gill wrote:


No, I've not tried playing with such things, although I've thought
about downsampling as simply a way of speeding up creating a gamut
from a large collection of images.
Yeah, I've been thinking about that too, though I've been wondering whether the easiest, most robust method is to create one image containing a single sample of every colour represented in the source images. Assuming 8-bit images, that would effectively cap the image size at 16.7 megapixels.
Am I right in assuming that tiffgamut doesn't currently do any weighting of the source pixels - that colours that appear frequently aren't treated with any more significance than colours that crop up rarely? 


The effect on the overall
gamut mapping may be very subjective and image dependent, I don't
known, and it would need some work to figure out what is reasonable.
I'm currently working on a (GTK-based and open source) GUI for the tiffgamut/icclink trick, with photographic output in mind - but it'd be trivial to wedge a blur or downsample filter into that for further experimentation.
Oh, that reminds me - I see in the latest beta a few of the utilities have been renamed - have you given any more thought to the possibility of resolving the filename clash between Argyll's and LCMS's icclink utilities? 

All the best,
--
Alastair M. Robinson

Other related posts: