Make Camera Profile with DCamProf

Caveat: This post, strictly defined, may not belong in the RT forum but I can find no better location. It was motivated ty a recent addition to RawPedia found here:

http://rawpedia.rawtherapee.com/How_to_create_DCP_color_profiles#Creating_DCP_profiles_using_DCamProf

If the moderator thinks it belongs elsewhere please feel free to move it.

I have long been interested in creating better camera profiles and was led to the DCamProf software created by Anders Torger. I must confess that, after spending several hours reading the documentation, I was more than a little discouraged by the complexity of this approach. Thus I was overjoyed to find the simple and straightforward instruction (to create a dual illuminant profile) found in the RawPedia article based directly on Anders Torger’s work.

After studying the list of command line entries at step #3 it became apparent that some modification is required for myself and others using a Windows based OS. This post resents my attempt to convert these instruction for Windows users, mostly in the hope that others, more knowledgeable than I, will examine my work and correct any errors.

To simplify the process I have first made the following additions to my OS:

** Append to the PATH environmental variable: a path to the \bin folder found in the Argyll software.
[e.g. D:\MyExe\ArgyllCMS\bin]

**Append to the PATH envronmental variable: a path to the file dcamprof.exe found in the DCamProf software.
[e.g. D:\MyExe\DCamProf]

**Create an environmental variable ARGLREF which supplies a path to the \ref folder in the Argyll software.
[e.g. ARGLREF=D:\MyExe\ArgyllCMS\ref]

With these additions in place the command line instructions now read:

scanin -v -dipn tungsten.tif ARGLREF\ColorCheckerPassport.cht
ARGLREF\ColorCheckerPassport.cie tungsten-diag.tif

scanin -v -dipn daylight.tif ARGLREF\ColorCheckerPassport.cht
ARGLREF\ColorCheckerPassport.cie daylight-diag.tif

dcamprof make-profile -i StdA tungsten.ti3 tungsten.json

dcamprof make-profile -i D50 -C daylight.ti3 daylight.json

dcamprof make-dcp -n “RT Camera Name” -n “RT Camera Name” -t acr -o neutral tungsten.json daylight.json “RT Camera Name.dcp”

The command prompt should be opened in the folder containing your tungsten.tif and daylight.tif files. Alternatively prepend the full path to these files to the filenames.

With apologies to the original author (to whom I extend a hearty thank you) I have made some minor changes:

In lines 1 & 2 I have substituted ColorCheckerPassport.cht taken from Argyll\ref in place of ColorChecker.cht from the same source.

In the same lines I have substituted ColorCheckerPassport.cie from Argyll\ref in place of cc24_ref.cie from DCamProf\data-examples.

Any comments or corrections gladly received.

I should also report that I have not yet given the altered program a trial run. Watch this space.

2 Likes

Clearly I am a rank beginner just exploring the possibilities of using DCamProf to create better camera profiles (DCP).

I was motivated by the relatively new page in RawPedia (cited in the first post above). After taking the first few steps in my journey, I have a question. Given the rather underwhelming response to the above post I am uncertain if I should continue to post in this thread but, finding no better spot, here goes:

I made a raw image of the Color Checker Passport, opened it in RT, Neutral profile. Cropped the image to just the color patch page of the Color Checker. Ensure the black patch is positioned at lower right. In Color Management tool: click “Save reference image for profiling”; apply white balance NOT checked. Resulting image file name: CCPP.tif.

Next, using the ArgyllCMS software, I proceeded to create a .val file (just as a sanity check). This file should contain the RGB values for each of the 24 color patches. Opened a command prompt in the folder containing the tif image.

scanin -o -dipn CCPP.tif ARGLREF\ColorCheckerHalfPassport.cht diag.tif

the resulting output files: diag.tif when opened showed perfect alignment with no errors or problems…
The file CCPP.val gave R, G & B values for each of the color patches

The PROBLEM: The RGB values shown by the file did not come even close to approximating the reference values for the ColorCheckerPassport derived from several online sources. Is this to be expected? If not, where have I gone wrong?

I’d disagree, this post has 132 views, which seems like quite a few to me. I’d personally like you to continue to share. It is these types of posts that’d I’d love to take, edit a bit more, and post to the main articles section :slight_smile:

@paperdigits

I’d disagree, this post has 132 views, which seems like quite a few to me. I’d personally like you to continue to share.

Thanks for your encouragement. Lots of views; yes. But I would really like some comment on the substance of the posts.