First of all i 'm working with a Fuji Finepix HS50 EXR

I'm doing an internship and my job is to collect photosite raw data to analyse phosphorescent paint comportment.

To do so I need to access the image before any process is done because it would alter our measurement.

I saw this post on dpreview: Access to "Bayer RAW" pre-processed image data

and it was pretty much what I was aiming for. ( I tried to follow instructions from the CHDK forums to use dcraw). After few searches I concluded that dcraw was the tool I needed to use because it was one of the few capable of reading the Finepix CFA. With RawTherapee I get a similar result as the post on dpreview but I doubt the software gives me an unaltered file.

Could someone explain how to get a pre-demosaicing image conserving all the information .

I'm currently using an alternative to dcraw with these parameters:

dcrawjdc.exe -v -r 1 1 1 1 -H 1 -o 0  -j -T -q 100

but unfortunately it got the wrong CFA array combination AF

  • 4
    I'm voting to close this question as off-topic because using a camera as a scientific measuring instrument is not related to photography. – Please Read My Profile May 11 '15 at 13:46
  • 1
    Ditto to what mattdm said, also because it is not clear how the results can be useful given the spectral response of the sensor is not determined. – Iliah Borg May 11 '15 at 14:16

As far as I understand your question, it seems to me that the document mode from the interpolation options should solve your problem, see man dcraw:

    Show the raw data as a grayscale image with no interpolation. 
    Good for photographing black-and-white documents. 
    Same as -d, but with the original unscaled pixel values. 
    Same as -D, but masked pixels are not cropped.

See also http://www.camerahacker.com/Digital/dcraw_by_example.shtml for an example.

I just tested this on the sample raw files from http://www.photographyblog.com/reviews/fujifilm_finepix_hs50exr_review/sample_images/ using

dcraw -v -D -j -T fujifilm_finepix_hs50exr_01.raf 

gives an output that looks as follows

bayer pattern

| improve this answer | |
  • First of all thanks for answering that quick ! I'm already using the interpolation options you're talking about (I think -D and -d are mutually exclusive). The result is a grayscale image and from here I do not know how to get RGB values (if that's possible) edit :Here is an example of output from dcraw link – Johny B. May 11 '15 at 13:52
  • 6
    @JohnyB. I'm afraid I don't understand your question then. Raw files before demosaicing are "grayscale" in the sense that each pixel only records luminance w.r.t. one single colour. This means, each pixel is either an R, G, or B value. Demosaicing then combines neighbouring pixels in order to obtain colour information, but in this stage you are already going "post-Bayer". If you want the "pre-Bayer" information, that's the image above. If you can clarify your question, I can give it another try. – godfatherofpolka May 11 '15 at 14:05
  • Thank's a lot sir, I was misunderstanding the content of the "grayscale" image. Since it's not the right place to post I will try to explain my need in another forum. Thank you again, have a good day – Johny B. May 12 '15 at 6:05

Not the answer you're looking for? Browse other questions tagged or ask your own question.