Re: color space of a scanned image
- From: "vonschwartzwalder" <vonschwartzwalder@xxxxxxx>
- Date: 23 Jun 2006 14:28:25 -0700
Wil,
With a color profile you are probably getting sRGB from the scanner. I
didn't say sRGB from JPEG, because I don't know how your JPEG decoder
works.
I can't really say much about how your system is set up. One thing you
could do is print a test pattern of color blocks, scan it, display the
scan on screen and see if the screen matches the printed test pattern.
If it does you are either very lucky or your entire system (printer,
screen and scanner) are using profiles. You won't get a perfect match,
though, because the screen gamut and printer gamuts are different
(printer is smaller).
As for the manual describing things: this is probably a consumer level
device. The technical writers don't include stuff like that because it
just generates support calls and/or confuses the user. The hardware is
likely much more capable than they tell you. A profile system is not
something everyday users care about, they just want the colors to be
'good'.
Once you know the words 'color space' you become a 'power user' and you
get to learn everything from somewhere else. Given today's technology,
if the above test works, you get sRGB.
How critical is your color measurement? Given you are using a consumer
level scanner you can probably just assume sRGB and convert to L*a*b*
from there. If that's not good enough you'll probably have to invest
in something a little better for color measurement.
duane
news.purdue.edu wrote:
Dear Duane,
Thanks for your reply.
When you said RGB, do you mean sRGB?
When a scanner is set up, the ICC profile is installed automatically and is
used when it scans, is that correct?
Is there any rule here how I can determine it? I did have checked the manual
and specification of the scanner, but didn't get anything related to color
space.
wil.
"vonschwartzwalder" <vonschwartzwalder@xxxxxxx> wrote in message
news:1151066186.296774.15360@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
There are two likely possibilities:
1) if you got a JPEG, it's YCbCr that is converted to RGB when decoded
2) any other file type on Windoze is RGB
If you used an ICC profile, you probably got sRGB, so converting to
L*a*b* is possible.
Is that what you are after, or is there a real question behind this
one?
duane
wil wrote:
Hi all,
I read a lot of materials on the web related to color management but
still could not answer this question.
I used a HP flatbed scanner connected to a Windows XP PC and created a
few scanned samples. I need to further manipulate them, and in
particular I want to perform a few experiments that require converting
the data into L*a*b* color space. Before doing this, I however don't
know what color space the image files are now in.
Anyone can help me to figure this out? Is there anything I can look
into to find this out? Is it related to the setting of the ICC profile
installed with the scanner?
I appreciate for your kind help. Thanks a lot.
wil.
.
- Follow-Ups:
- Re: color space of a scanned image
- From: news.purdue.edu
- Re: color space of a scanned image
- References:
- color space of a scanned image
- From: wil
- Re: color space of a scanned image
- From: vonschwartzwalder
- Re: color space of a scanned image
- From: news.purdue.edu
- color space of a scanned image
- Prev by Date: open source 3d triangulation or mesh from point cloud?
- Next by Date: Re: Gaussian luminance distribution
- Previous by thread: Re: color space of a scanned image
- Next by thread: Re: color space of a scanned image
- Index(es):
Relevant Pages
|
|