[Digikam-devel] Colormanagement

[Peter Heckert]

Hello,

have some problems with colormanagement.

I am trying to use linear RGB as a working space.
For this purpose I am using this (homemade) colorprofile as workingspace:
<http://home.arcor.de/peter.heckert/raw/RawSharpen/sRGBG1.icc>

This is a profile that uses sRGB primary colors, but instead of
Gamma=2.2 it uses Gamma = 1.0.

My imaging software has full colormanagement and is fully 16-Bit
capable. (Photoline32 for Windows, http://www.pl32.com)

In this application I have two choices: I can /assign/ a colorspace to
an image or I can /convert/ an image into another colorspace.

In the first case,"assign",  the image data is not modified, only its
interpretation is modified and so the visual look will change.
In the second case, "convert", the image data is converted into another
colorspace, the data is modified, but the display stays unchanged,
because also  the interpretation of image data changes.

Normally I do this for my RAW Images: I load the image with Gamma=1.0.
The image will look dark then.

Then I /assign/ sRGBG1.icc to the image. Then the image looks rather
fine. (Some contrast improvements and else are necessary anyway)

The image is not converted in any way. The image processor will adjust
image's *displaybuffer* accordingly, it doesnt modify the image itself.

That means, all image manipulations in follow will be done in linear RGB
space. All filters and all blending modes work in linear RGB.

That is how I understand the word "workingspace".

Ok, here is another trick:
Sharpening works much better in linear RGB, because sharpening halos are
reduced. Denoising works better, because local noise amplitude is
independent from local brightness in linear RGB space.

Therefore I often convert my JPG's to 16 Bit and then I /convert/ to
sRGBG1.icc before I continue with image manipulations.

(I convert to 16 bit, because linear RGB cannot be used with 8 bit
resolution, there would be too much loss of information)

I tried to do the same trick in digikam.
In digicam I cannot convert colorspaces. I can choose another working
space. However, when I choose another working space, then the image's
visual onscreen appearence is changed and so far I understand it, this
should not happen.

Ok, now I am unsure: Is this a bug or do I missunderstand
colormanagement philosophy or is colormanagegement implementation still
unfinished?

greetings,

Peter
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-2]

Le Dimanche 02 Avril 2006 04:35 AM, Peter Heckert a écrit :

> Hello,
>
> have some problems with colormanagement.
>
> I am trying to use linear RGB as a working space.
> For this purpose I am using this (homemade) colorprofile as workingspace:
> <http://home.arcor.de/peter.heckert/raw/RawSharpen/sRGBG1.icc>
>
> This is a profile that uses sRGB primary colors, but instead of
> Gamma=2.2 it uses Gamma = 1.0.
>
> My imaging software has full colormanagement and is fully 16-Bit
> capable. (Photoline32 for Windows, http://www.pl32.com)
>
> In this application I have two choices: I can /assign/ a colorspace to
> an image or I can /convert/ an image into another colorspace.
>
> In the first case,"assign",  the image data is not modified, only its
> interpretation is modified and so the visual look will change.
> In the second case, "convert", the image data is converted into another
> colorspace, the data is modified, but the display stays unchanged,
> because also  the interpretation of image data changes.
>
> Normally I do this for my RAW Images: I load the image with Gamma=1.0.
> The image will look dark then.
>
> Then I /assign/ sRGBG1.icc to the image. Then the image looks rather
> fine. (Some contrast improvements and else are necessary anyway)
>
> The image is not converted in any way. The image processor will adjust
> image's *displaybuffer* accordingly, it doesnt modify the image itself.
>
> That means, all image manipulations in follow will be done in linear RGB
> space. All filters and all blending modes work in linear RGB.
>
> That is how I understand the word "workingspace".
>
> Ok, here is another trick:
> Sharpening works much better in linear RGB, because sharpening halos are
> reduced. Denoising works better, because local noise amplitude is
> independent from local brightness in linear RGB space.
>
> Therefore I often convert my JPG's to 16 Bit and then I /convert/ to
> sRGBG1.icc before I continue with image manipulations.
>
> (I convert to 16 bit, because linear RGB cannot be used with 8 bit
> resolution, there would be too much loss of information)
>
> I tried to do the same trick in digikam.
> In digicam I cannot convert colorspaces. I can choose another working
> space. However, when I choose another working space, then the image's
> visual onscreen appearence is changed and so far I understand it, this
> should not happen.
>
> Ok, now I am unsure: Is this a bug or do I missunderstand
> colormanagement philosophy or is colormanagegement implementation still
> unfinished?

screen ICC color correction is not yet fully implemented. It's must done at
0.9.0. Paco, another digikam developer working on actually.

Tak a look in Paco blog : http://www.digikam.org/?q=blog/5

Gilles
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[F.J.Cruz]

---- Peter Heckert <[hidden email]> escribió:

> Hello,
>
> have some problems with colormanagement.
>
> I am trying to use linear RGB as a working space.
> For this purpose I am using this (homemade) colorprofile as workingspace:
> <http://home.arcor.de/peter.heckert/raw/RawSharpen/sRGBG1.icc>
>
> This is a profile that uses sRGB primary colors, but instead of
> Gamma=2.2 it uses Gamma = 1.0.
>
> My imaging software has full colormanagement and is fully 16-Bit
> capable. (Photoline32 for Windows, http://www.pl32.com)
>
> In this application I have two choices: I can /assign/ a colorspace to
> an image or I can /convert/ an image into another colorspace.
>
> In the first case,"assign",  the image data is not modified, only its
> interpretation is modified and so the visual look will change.
> In the second case, "convert", the image data is converted into another
> colorspace, the data is modified, but the display stays unchanged,
> because also  the interpretation of image data changes.
>
> Normally I do this for my RAW Images: I load the image with Gamma=1.0.
> The image will look dark then.
>
> Then I /assign/ sRGBG1.icc to the image. Then the image looks rather
> fine. (Some contrast improvements and else are necessary anyway)
>
> The image is not converted in any way. The image processor will adjust
> image's *displaybuffer* accordingly, it doesnt modify the image itself.
>
> That means, all image manipulations in follow will be done in linear RGB
> space. All filters and all blending modes work in linear RGB.
>
> That is how I understand the word "workingspace".
>
> Ok, here is another trick:
> Sharpening works much better in linear RGB, because sharpening halos are
> reduced. Denoising works better, because local noise amplitude is
> independent from local brightness in linear RGB space.
>
> Therefore I often convert my JPG's to 16 Bit and then I /convert/ to
> sRGBG1.icc before I continue with image manipulations.
>
> (I convert to 16 bit, because linear RGB cannot be used with 8 bit
> resolution, there would be too much loss of information)
>
> I tried to do the same trick in digikam.
> In digicam I cannot convert colorspaces. I can choose another working
> space. However, when I choose another working space, then the image's
> visual onscreen appearence is changed and so far I understand it, this
> should not happen.
>
> Ok, now I am unsure: Is this a bug or do I missunderstand
> colormanagement philosophy or is colormanagegement implementation still
> unfinished?
>
> greetings,
>
> Peter

Hi Peter,

My name is Paco Cruz and I'm the developer for all de color management stuff.

In this moment, digiKam has a basic color management support that I'm going to explaint to you:

digiKam can convert from input device color profile to working or editing color profile and to output one, this is, we can convert an image from camera/scanner profile to other one suitable for image editing (e.g. Adobe RGB) or to printer color profile.

Actually, I'm working to provide digiKam with a "managed view" option, using the display (monitor) color profile.

digiKam doesn't support "assign" option, only "convert" one. As far I can remember now, with the first option (in photoshop) the image you see in the monitor stay unchanged and with the second one, the image is changed, but I'm not sure about this.

In any way, I'm not an expert in color management and for a long time I've been asking for help to get digikam a fully color managed  app but looks like nobody in developer team or mailing lists have experience on this stuff, so if you have this experience, I'll be glad to hear what you can say me :-).

PD: Excuse my poor english ;-).

Paco Cruz.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

Am Montag, 3. April 2006 12:25 schrieb [hidden email]:

> Actually, I'm working to provide digiKam with a "managed view" option,
> using the display (monitor) color profile.

BTW, how does a managed view works in a multi-monitor setup? :-)
Does anybody know how other systems handle CMS in a Xinerama-like set-up?

Bye

  Thorsten

_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Peter Heckert]

Hi Paco,

[hidden email] wrote:
 > Hi Peter,
 >
 > My name is Paco Cruz and I'm the developer for all de color
 > management stuff.
 >
 > In this moment, digiKam has a basic color management support that I'm
 > going to explaint to you: digiKam can convert from input device color
 > profile to working or editing color profile and to output one, this
 > is, we can convert an image from camera/scanner profile to other one
 > suitable for image editing (e.g. Adobe RGB) or to printer color
 > profile. Actually, I'm working to provide digiKam with a "managed
 > view" option, using the display (monitor) color profile.
 >
Yes, please forget all, what I wrote.
I am still exploring digkam and after writing this I discovered the
global settings for colormanagement.
Ok, I was not able to use it, because it crashes or doesnt respond when
I set the input profile. So I believe, it is still unfinished code.

 > digiKam doesn't support "assign" option, only "convert" one. As far I
 > can remember now, with the first option (in photoshop) he image you
 > see in the monitor stay unchanged and with the second one,
 > the image is changed, but I'm not sure about this.

I believe, the "Assign" operation in photoline32 is equal to setting an
input profile in digikam. It's just another name.

I understand it this way: When an image is loaded, that has no profile
embedded then the app "assigns" a default color profile and if this
doesnt match, then the user must assign a profile manually (in pl32) or
set the input profile manually (in digikam).

 >
 > In any way, I'm not an expert in color management and for a long time
 > I've been asking for help to get digikam a fully color managed  app
 > but looks like nobody in developer team or mailing lists have
 > experience on this stuff, so if you have this experience, I'll be
 > glad to hear what you can say me :-) .
 >

I am not an expert at all. However, pl32, while it is a windows app, it
is a very nice program, it has two developers and they are very helpful
to users in their discussion board and they explained to me what to do
in order to edit images in linear colorspace.

It is commercial shareware, but you can download it at
<http:www.pl32.com> and use it forever if you tolerate a nag screen at
startup. If you want you can try it, it is simpler than photoshop and
that means it is less confusing ;-)
So far I know, they also use lcms.

I find this site very helpful and informative : <http://www.aim-dtp.net>

 > PD: Excuse my poor english ;-) .
 > It is at least better than my english ;-)
 > Paco Cruz.
 >
Ok, be warned, I have a special wish for colormanagement. I already told
this to pl32 developers, but they did not implement this until now.

Here it is:

When I load an 8 Bit image, then this is usually an sRGB image so I have
sRGB as my default profile.

However, when I load a 16-Bit image, then this is usually a raw file or
a selfmade tiff file and this is in linear colorspace (sRGBG1.icc).

That's my working philosophy, other people may have another philosophy.

Anyway, there are reasons to use another input space and working space
for 8 bit images different from the working space for 16 bit images.

So it would be my wish for future versions to have different default
colormanagement settings for 8-bit and for 16-bit images.
When an image is converted from 8 bit to 16 bit or vice versa then it
would be fine to have a default colorspace conversion connected with
this operation. ;-)

Ok, I understand there is a heavy time pressure for the next beta and so
please take this as a wish for future versions, not for the next version.

P.S. I am a very beginner, presently I trie to learn how to debug a
digikam plugin.
Still trying, only partial success, still learning...


Peter

_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[F.J.Cruz]

El Lunes, 3 de Abril de 2006 15:36, Thorsten Schnebeck escribió:

> Am Montag, 3. April 2006 12:25 schrieb [hidden email]:
> > Actually, I'm working to provide digiKam with a "managed view" option,
> > using the display (monitor) color profile.
>
> BTW, how does a managed view works in a multi-monitor setup? :-)
> Does anybody know how other systems handle CMS in a Xinerama-like set-up?
>
> Bye
>
>   Thorsten
>

Hi Thorsten,

If the two monitors are the same model, then is possible there is no problem,
but if monitors are differents models or from differents
manufacturers.....you have to use one of them to "see" the image and the
other one to show the tools, because digiKam only can use one profile at
once.

Of course you have to select the profile from the "to see" monitor :-).

Paco
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-2]

Le Lundi 03 Avril 2006 11:20 PM, F.J.Cruz a écrit :

> El Lunes, 3 de Abril de 2006 15:36, Thorsten Schnebeck escribió:
> > Am Montag, 3. April 2006 12:25 schrieb [hidden email]:
> > > Actually, I'm working to provide digiKam with a "managed view" option,
> > > using the display (monitor) color profile.
> >
> > BTW, how does a managed view works in a multi-monitor setup? :-)
> > Does anybody know how other systems handle CMS in a Xinerama-like set-up?
> >
> > Bye
> >
> >   Thorsten
>
> Hi Thorsten,
>
> If the two monitors are the same model, then is possible there is no
> problem, but if monitors are differents models or from differents
> manufacturers.....you have to use one of them to "see" the image and the
> other one to show the tools, because digiKam only can use one profile at
> once.
>
> Of course you have to select the profile from the "to see" monitor :-).

I second Paco about this point. In my office, i'm using Xinerama with 3
monitors. I'm working like this...

Gilles
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel