[poppler] Color Management

Albert Astals Cid aacid at kde.org
Wed Jan 7 08:56:10 PST 2009 

A Divendres 12 Desembre 2008, Koji Otani va escriure:
> Very sorry, I attached a wrong file to the previous mail.
> I attached a correct file to this mail.

I've commited the patch as i didn't find anything bad at first sight, probably 
setupColorProfiles should be tweaked in the future to allow changing the 
display profile on the fly, but that's not a huge problem since none of our 
consumers uses that at the moment :D

Albert

> ---------
> Koji Otani.
>
> From: Koji Otani <sho at bbr.jp>
> Subject: Re: [poppler] Color Management
> Date: Fri, 12 Dec 2008 14:16:40 +0900 (JST)
> Message-ID: <20081212.141640.112579191.sho at bbr.jp>
>
> sho> I'm Koji Otani.
> sho>
> sho> I found a mistake in my patch.
> sho> (one #ifdef --> #if)
> sho> So, in case not using CMS, compiling error occurs.
> sho>
> sho> I attached fixed patch.
> sho>
> sho>
> sho>
> sho>
> sho>
> sho> From: Koji Otani <sho at bbr.jp>
> sho> Subject: Re: [poppler] Color Management
> sho> Date: Thu, 27 Nov 2008 17:57:37 +0900 (JST)
> sho> Message-ID: <20081127.175737.52202455.sho at bbr.jp>
> sho>
> sho> sho> Hi, All
> sho> sho>
> sho> sho> I made a patch to add Color Management to poppler.
> sho> sho> I attached it to this mail.
> sho> sho> This patch is one for current source in the GIT repository.
> sho> sho>
> sho> sho> I'm not a CM expert. And this patch doesn't resolve all problems.
> sho> sho> But, I hope to make some progress.
> sho> sho> Please try this.
> sho> sho> ------------
> sho> sho> Koji Otani.
> sho> sho>
> sho> sho> --------------
> sho> sho> This patch adds Color Management feature to polper:
> sho> sho> (1) Handling ICCBasedColorSpace (both V2 and V4 profiles)
> sho> sho> (2) Handling CalRGB and CalGray
> sho> sho>
> sho> sho> This doesn't handle Rendering Intent.
> sho> sho>     This always renders with relative colorimetric intent
> (default in PDF). sho> sho> This doesn't handle transparency.
> sho> sho>
> sho> sho> This uses Little cms library, so when compiling this, lcms-devel
> package is sho> sho>  needed. if lcms-devel package is not installed,
> disable this feature. sho> sho>
> sho> sho> In default, ICCBased uses builin Standard RGB,
> sho> sho>    Lab, CalRGB and CalGray don't use profile.
> sho> sho> If you want to use other profile, put profile file to
> sho> sho>   $HOME/.xpdf/ColorProfiles/display.icc
> sho> sho>     or
> sho> sho>   /usr/share/poppler/ColorProfiles/display.icc
> sho> sho> This profile is used by ICCBased, Lab, CalRGB, CalGray.
> sho> sho> You can also change display profile from your application.
> sho> sho>   Before calling PDFDoc::displayPage, call
> sho> sho>    GfxColorSpace::setDisplayProfileName
> sho> sho>     or
> sho> sho>    GfxColorSpace::setDisplayProfile
> sho> sho>
> sho> sho> static void setDisplayProfile(cmsHPROFILE displayProfileA)
> sho> sho>     set profile handle
> sho> sho> static void setDisplayProfileName(GooString *name)
> sho> sho>     set profile file name
> sho> sho>       If file name started with '/', it is handled as absolute
> path name. sho> sho>       Otherwise, it is handled as relative path to
> sho> sho>          $HOME/.xpdf/ColorProfiles/
> sho> sho> 	 and
> sho> sho>          /usr/share/poppler/ColorProfiles/
> sho> sho> -----------
> sho> sho>
> sho> sho>
> sho> sho> From: "Hal V. Engel" <hvengel at astound.net>
> sho> sho> Subject: Re: [poppler] Color Management
> sho> sho> Date: Thu, 21 Aug 2008 18:51:59 -0700
> sho> sho> Message-ID: <200808211852.00098.hvengel at astound.net>
> sho> sho>
> sho> sho> hvengel> On Thursday 07 August 2008 06:58:13 pm Leonard Rosenthol
> wrote: sho> sho> hvengel> > Correct.  You'd need to make MAJOR changes to
> the rendering sho> sho> hvengel> > architecture of Poppler to do proper
> color management - especially sho> sho> hvengel> > where transparency is
> involved.
> sho> sho> hvengel> >
> sho> sho> hvengel> > But it would be a very good thing...
> sho> sho> hvengel> >
> sho> sho> hvengel> > Leonard
> sho> sho> hvengel>
> sho> sho> hvengel> (This is kind of long please bear with me).
> sho> sho> hvengel>
> sho> sho> hvengel> I have been looking at the PDF spec. and at the poppler
> code.  Leonard is sho> sho> hvengel> correct that the code in it's present
> form is not designed to support color sho> sho> hvengel> management and it
> will require major changes.
> sho> sho> hvengel>
> sho> sho> hvengel> Some exmples:
> sho> sho> hvengel>
> sho> sho> hvengel> The PDF spec calls for all CIE based colorspaces (Lab,
> calRGB, calCMYK and sho> sho> hvengel> ICCBased) to undergo the following
> transformations: sho> sho> hvengel>
> sho> sho> hvengel> original color space -> XYZ -> output color space
> sho> sho> hvengel>
> sho> sho> hvengel> This implies that there is clear separation between the
> first conversion and sho> sho> hvengel> the second conversion and that one
> set of routines can handle the second sho> sho> hvengel> conversion for all
> CIE based color transforms.
> sho> sho> hvengel>
> sho> sho> hvengel> It also implies that there is some way for calling
> applications to specifiy a sho> sho> hvengel> CIE based output color space
> such as an ICC profile and related information sho> sho> hvengel>
> (rendering intents, black point comp. and output channel depth).  This is
> sho> sho> hvengel> currently not possible.
> sho> sho> hvengel>
> sho> sho> hvengel> The code as it exists only does a direct conversion to
> XYZ in one location sho> sho> hvengel> GfxLabColorSpace::getRGB() and then
> does a second conversion to an arbitrary sho> sho> hvengel> RGB** color
> space in the same function.  None of the other CIE based color sho> sho>
> hvengel> space conversions produce an intermeadiate XYZ conversion and of
> course there sho> sho> hvengel> are no generic XYZ to output color space
> routines. sho> sho> hvengel>
> sho> sho> hvengel> The gray and cmyk code for Lab conversions uses the
> getRGB() function and then sho> sho> hvengel> does some kind of generic
> conversion using the RGB values.  Since these RGB sho> sho> hvengel> values
> are in an arbitrary RGB** color space the conversion to gray and CMYK sho>
> sho> hvengel> is also arbitrary.
> sho> sho> hvengel>
> sho> sho> hvengel> CalGray does not do gamma compensation in getGray() as
> called for in the PDF sho> sho> hvengel> spec.  In addition the CalGray
> functions only support 8 bit depth.  What sho> sho> hvengel> happens if
> there is a 16bit/channel gray image in a PDF file or if a user sho> sho>
> hvengel> wants 16 bit/channel output for his/her printer? sho> sho>
> hvengel>
> sho> sho> hvengel> CalRGB passes RGB values directly through to the output
> and makes no attempt sho> sho> hvengel> to convert these into an
> intermeadiate XYZ color space or into the actual sho> sho> hvengel> output
> colorspace.  It also does not apply a gamma correction to the data as sho>
> sho> hvengel> per the PDF spec.
> sho> sho> hvengel>
> sho> sho> hvengel> End examples:
> sho> sho> hvengel>
> sho> sho> hvengel> I don't think that I am pointing out anything new to
> most on this list.  When sho> sho> hvengel> I first started to look at the
> code I was hoping that it would not be too sho> sho> hvengel> difficult to
> find the places where CM hooks could be put in place and that sho> sho>
> hvengel> perhaps I could spend a few days putting together a set of patches
> that would sho> sho> hvengel> provide a starting place.  But it appears
> that the code needs significant sho> sho> hvengel> restructuring in order
> to even start doing the actaul color management sho> sho> hvengel> specific
> work.  Fortunately the PDF specification has enough detail that it sho>
> sho> hvengel> should be possible for the poppler team to do much of the
> restructuring work sho> sho> hvengel> without too much involvement from
> someone with color management expertise. sho> sho> hvengel>
> sho> sho> hvengel> Mostly what is needed is:
> sho> sho> hvengel>
> sho> sho> hvengel> 1. An API for applications to specify the output color
> space, rendering sho> sho> hvengel> intent, black point compensation and
> channel depth for a document.  These are sho> sho> hvengel> needed to
> correctly setup the XYZ to output color space transform. sho> sho> hvengel>
> sho> sho> hvengel> 2. The CIE related code (calRGB, calGray, Lab and
> ICCBased) needs to be sho> sho> hvengel> restructured so that it has a
> clean division between producing the sho> sho> hvengel> intermeadeate XYZ
> values and the code that does the XYZ to output color space sho> sho>
> hvengel> conversion.  See the diagram on pages 238 and 239 of the version
> 1.7 PDF sho> sho> hvengel> Reference.
> sho> sho> hvengel>
> sho> sho> hvengel> Initially the ICCBased code could be left alone and
> calRGB, calGray and Lab sho> sho> hvengel> would be setup to convert to XYZ
> but would still do the current simplistic sho> sho> hvengel> conversions to
> the output color space (IE. these would all look some what like sho> sho>
> hvengel> GfxLabColorSpace::getRGB() & friends) .
> sho> sho> hvengel>
> sho> sho> hvengel> Then the XYZ to output color space routines would be
> added and the calRGB, sho> sho> hvengel> calGray and Lab routines would
> call them to handle the output conversion in a sho> sho> hvengel> more
> correct way.  It is at this point where the code would start making use
> sho> sho> hvengel> of a CMS like LCMS as well as the API that was created
> in #1.  This sho> sho> hvengel> functionality is documented in the diagram
> on page 239 of the PDF version 1.7 sho> sho> hvengel> spec as "Conversion
> from CIE-based to device color space (not specified by sho> sho> hvengel>
> PDF)".
> sho> sho> hvengel>
> sho> sho> hvengel> 3. At this point setting up the ICCBased routines to
> create XYZ intermeadiate sho> sho> hvengel> results would be added.  Once
> this is in place poppler would have a sho> sho> hvengel> functioning color
> managed system for all of the CIE based color space types. sho> sho>
> hvengel>
> sho> sho> hvengel> I have not looked at the PDF specs for "Special color
> spaces" such as sho> sho> hvengel> Separation, DeviceN, Indexed and Pattern
> so I do not know what implications sho> sho> hvengel> there are for these
> in a color managed system.  I have also not looked at sho> sho> hvengel>
> transparancy issues but I didn't see much in the spec that related to
> ICCBased sho> sho> hvengel> objects other than that it was nessassary to
> have to AtoB and BtoA tables in sho> sho> hvengel> the profiles to support
> this and that all blending must be either in device sho> sho> hvengel>
> space or in CIE space and that the spec advises that CIE is prefered. sho>
> sho> hvengel>
> sho> sho> hvengel> Many of you are probably asking "Why should we give a
> rats behind about this?" sho> sho> hvengel> The reason I am looking at this
> is that the printing community is in the sho> sho> hvengel> process of
> converting from PostScript to PDF as the standard document type for sho>
> sho> hvengel> printing for *nix systems.  Because of this they are in the
> process of writing sho> sho> hvengel> a new pdftoraster filter for CUPS as
> well as putting together a PDF based sho> sho> hvengel> "Common Printing
> Dialog" (this is a Google Summer of Code project).  I asked sho> sho>
> hvengel> on the printing email lists about color management in the new
> printing sho> sho> hvengel> workflow.  Specifcally did the new pdftoraster
> filter have CM support?  I sho> sho> hvengel> didn't get an answer and so I
> found the code and had a look.  Guess what (most sho> sho> hvengel> of you
> probably know this) it is using poppler and as a result it does not sho>
> sho> hvengel> support color management.
> sho> sho> hvengel>
> sho> sho> hvengel> Printing is an important piece of the infastructure in
> general and sho> sho> hvengel> particularly for anyone doing color critical
> work.  It is currently badly sho> sho> hvengel> broken with respect to
> color managemnt and it appears that until poppler has sho> sho> hvengel> CM
> support or another similar library with CM support becomes available it
> sho> sho> hvengel> will remain broken.  So now you know what the story is
> and why this is sho> sho> hvengel> important.
> sho> sho> hvengel>
> sho> sho> hvengel> I have color management expertise (I am the maintainer
> of LProf) and I have sho> sho> hvengel> connections to the open source
> color management community where there are many sho> sho> hvengel> others
> with CM expertise some of them with way more expertise than I have.  As
> sho> sho> hvengel> a community we have been working with the printing
> community, Xorg and the sho> sho> hvengel> DE's trying to get the missing
> CM components in place. sho> sho> hvengel>
> sho> sho> hvengel> There has been major progress with color management on
> monitors and we now sho> sho> hvengel> have calibration and profiling
> software for these devices as well as support sho> sho> hvengel> in XOrg
> (with more on the way) and work is underway to extend this to include sho>
> sho> hvengel> better user tools in KDE and GNOME.  We also have good open
> source tools for sho> sho> hvengel> profiling input devices like cameras
> and scanners and these profiles are now sho> sho> hvengel> well supported
> by things like XSane, UFRAW and other software in this area. sho> sho>
> hvengel>
> sho> sho> hvengel> We also have high quality profiling software for output
> devices like printers. sho> sho> hvengel> One of the major things that is
> missing is a viable color managed path for sho> sho> hvengel> printing. 
> That is I can create very high quality profiles for my printers but sho>
> sho> hvengel> using them with the existing printing tools is at best
> difficult even for sho> sho> hvengel> someone who really understands CM and
> nearly imposible for a normal user.   If sho> sho> hvengel> the CUPS
> pdftoraster filter had CM support much of the printing part of this sho>
> sho> hvengel> would start falling in place and would become accessable to a
> much wider sho> sho> hvengel> audiance.  So from the point of view of the
> CM community poppler is now a very sho> sho> hvengel> important piece of
> software.
> sho> sho> hvengel>
> sho> sho> hvengel> Hal
> sho> sho> hvengel>
> sho> sho> hvengel> PS For those interested this web page from the ICC has a
> link to a PDF file sho> sho> hvengel> from Adobe that demonstrates CM or a
> lack of CM as the case may be: sho> sho> hvengel>
> sho> sho> hvengel>   http://www.color.org/version4html.xalter
> sho> sho> hvengel>
> sho> sho> hvengel>
> sho> sho> hvengel> **It looks like it might end up being sort of sRGB since
> it uses a matrix sho> sho> hvengel> conversion and then a gamma conversion
> that is too simple to give actual sRGB sho> sho> hvengel> results since
> sRGB has a compound gamma curve.
> sho> sho> hvengel>

More information about the poppler mailing list