Re: cr2 by raw2lum (fwd)

[Jan Hollan <jhollan@amper....muni.cz>]

Wannes and Roxana,

in the cr2proc.sh there was still another error,  namely the -i parameter 
in the first call to dcraw. This parameter, whose output is exif 
info, cannot be combined with parameters producing pgm output.

The corrected script is within
  http://amper.ped.muni.cz/light/luminance/scripts
 directory, at the top (cr2.zip)

The good news is, that after installing all needed software to a windoze 
computer (I never did anything like that before, changing windoze to linux 
in a way...), the scripts work OK. I am not sure about paths, so I put 
dcraw.exe and raw2lum.exe (raw2lum134.cne.exe can be renamed this way 
there) into the same directory as the *.cr2 files were in.

The list of the software which had to be installed is at the end of the
edited former letter, as cr2.txt within the cr2.zip.

Hope you'll master to make your own scripts, which would suit real data 
processing. A typical simplest one has two parameters, an image file name 
and a darframe name.


best, jenik

PS
 I've added a hyperlink to the README.html file at the end of the
  http://amper.ped.muni.cz/light/luminance/
 directory, leading to exiftool. Dave Coffin has links on his page
leading to dcraw for MSwindows. Maybe, a setup programme exists for it 
too.


On Sat, 5 Sep 2009, Jan Hollan wrote:

> Hi Wannes,
>
> in spite of saying "raw", my programme does not work with raw formats 
> (there are too many, having a complicated structure)
> just with raw Bayer matrix data. The input file has to be a *.pgm
> produced by dcraw. Pgm is the most simple image format possible.
>
> There should be some info of how to work with dcraw in windoze as well, 
> on my page http://amper.ped.muni.cz/light/luminance
>
> A student from Argentina had a similar problem recently, you might learn 
> perhaps something from the text and attachment I sent to her.
>
> Otherwise, send me a reasonable image with a proper darkframe (taken
> with a lid and the same exposure settings), I'll test if the new
> 134 version would work OK (don't know the new Canon resolutions etc.,
> perhaps I should enlarge field dimensions).
>
> best regards,
> jenik