Udi Fuchs schrieb:
Hi,
I wrote a small tool that uses DCRaw to read raw images from digital cameras, and uses LCMS for some basic color management. You can check it out at: http://ufraw.sourceforge.net/
I'm trying to keep the user interface simple, but it is a bit problematic. The problem is that in some cases the raw data has to be manipulated BEFORE applying the color transformation.
For example, the profile for the Nikon D70 (it can be downloaded from my site) requires a gamma=0.45 curve, this is a naive gamma curve of pow(x,0.45). Other profiles seem to require a gamma curve which is linear at the beginning. And some profiles are "linear" in the sense, that they can be applied to the raw data directly.
Hi Udi,
the basic concept of device profiles is to characterize the device (as it is), and not vice versa, to manipulate the image returned by an input device to fit with a given profile. If you only get correct results, if you preprocess the dcraw output before feeding it into the profile, then in fact you are not using the right profile for this task, and you should use a different profile instead, which is appropriate for your task.
If you create an input profile for dcraw processed Nikon D70 raw images, then this profile should not require any additional preprocessing.
But of course, if you just use an arbitrary profile you downloaded somewhere, why do you assume that this profile would fit at all with the output generated by dcraw? (and obviously, according to your statements, it does not in any case ...)
The point is, you basically need to use an input profile which was explicitly created for the particular camera IN CONJUNCTION WITH DCRAW, and not a profile which was e.g. created for the Nikon D70, for use in conjunction with Nikon software.
I suspect, your major problem is that you don't have the opportunity to create custom profiles for a large variety of cameras (for use in conjunction with dcraw) yourself, since I guess you don't have access to all these cameras, right?
But otherwise, IMO the only thing you can do, is what you're already doing: I.e. to GUESS what's the difference between the output of dcraw and the expected input of the (unknown) profile you're using.
Of course, once you have made a reasonable guess for the preprocessing transformation, then you could create a new profile, which combines both, the preprocessing and the original profile, so that you don't need to apply the preprocessing explicitly, but only need to apply the new profile.
Regards, Gerhard
Does someone know if the information on the required "preprocessing" is embedded somewhere in the ICC profile?
I'm amateur at color management, so any other comments on the way I apply the profiles will be welcome,
Udi
------------------------------------------------------- This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting Tool for open source databases. Create drag-&-drop reports. Save time by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc. Download a FREE copy at http://www.intelliview.com/go/osdn_nl _______________________________________________ Lcms-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/lcms-user
------------------------------------------------------- This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting Tool for open source databases. Create drag-&-drop reports. Save time by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc. Download a FREE copy at http://www.intelliview.com/go/osdn_nl _______________________________________________ Lcms-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/lcms-user
