Re: HP PhotoSmart R030 vs. SANE 1.01

abel deuring (a.deuring@satzbau-gmbh.de)
Wed, 28 Apr 1999 12:08:30 +0200

Russ Price wrote:
>
> I have one of these, and I've discovered that Kodachrome slides don't
> look too good when scanned. Even at 300% exposure with high levels of
> gamma correction, they still come out with a purplish tint. Are there
> any things I can reasonably do to improve this? The slides come out fine
> when using the windoze scanning software.

This points to a general problem of Sane: At the time, Sane has
no mechanisms for color correction. This is especially a problem
for slide scanners, because the color representation of negatives
is very different from the one of slides. But color correction
can also be useful for flatbed scanners.

The most easy solution for this is of course, to declare color
correction to be an issue for image processing programs like Gimp,
but, on the other hand, scanimage for example is very useful for
batch scans - and having batch scans color corrected, would be
fine. And the ICC standard for example has the concept of an
"abstract" representation of color data: It is up to the programs
for data aquisition to convert the data from the device color space
into the "abstract" color space, and it's the task of display /
output programs to convert the color represention into the output
device's color space.

Unfortunately, I don't have enough time to think much about how to
integrate color correction and / or calibration into Sane (not to
speak of implementing it...), but perhaps somebody else is
interested in this topic.

The ICC standard (http://www.color.org/index.html) is quite
powerful -- and it is, as far as I know, an open standard. So,
what about integrating ICC into Sane?

Abel

--
Source code, list archive, and docs: http://www.mostang.com/sane/
To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com