5 Replies Latest reply: Apr 26, 2012 4:44 AM by Noel Carboni RSS

    How to open with technical flat calibration in 7.0

    q38

      In earlier versions, Camera Raw gave an option to set "Blacks" = "Brightness" = Contrast" = "0", and set a straight tone curve.  With color space set to Adobe RGB, this opens with EXACTLY a calibrated power of 2.2.  This gives a very flat looking image with no "S" curve that no one would use for beauty, but when used for SCIENTIFIC applications, I can raise the 16-bit image to the 2.2th power in Photoshop and have EXACTLY a linear correspondence between photon count and pixel count.  This lets me add photons, divide images to normalize, make exact pixel by pixel measurements, and a lot of things us scientists like to do unrelated to making pictures beautiful.

       

      Camera Raw 7.0 changed the sliders, and after hours of scientific experimenting I can not find a combination that opens images that are technically exactly flat (gamma 2.2 of course) without deviations to the straight curve.

       

      I do weddings occasionally, science mostly, and Photoshop is my go-to tool for both.  Thanks in advance for your help advancing the cause of science.

        • 1. Re: How to open with technical flat calibration in 7.0
          Jeff Schewe Community Member

          q38 wrote:

           

          Camera Raw 7.0 changed the sliders, and after hours of scientific experimenting I can not find a combination that opens images that are technically exactly flat (gamma 2.2 of course) without deviations to the straight curve.

           

          You can't...but what you CAN do is create a custom DNG Profile in DNG Profile Editor that has a flat, linear curve and use that profile as the basis of a no curve starting point. Also note that the new PV 2012 does some auto highlight recovery and the image adjustments are based on image adpative range and end point settings.

          • 2. Re: How to open with technical flat calibration in 7.0
            Noel Carboni Community Member

            q38 wrote:

             

            Camera Raw 7.0 changed the sliders, and after hours of scientific experimenting I can not find a combination that opens images that are technically exactly flat (gamma 2.2 of course) without deviations to the straight curve.

             

            Is this true if you choose the PV2010 development process?  I had thought that was supposed to still be identical to what it was before.

             

            PV.jpg

             

            Nothing says you HAVE to use PV2012 to develop your scientific images.  You can set the development process into the Camera Raw defaults or create a Preset.

             

            -Noel

            • 3. Re: How to open with technical flat calibration in 7.0
              areohbee Community Member

              There's a lengthy discussion about this in the LIghtroom forum:

               

              http://forums.adobe.com/message/4289492#4289492

              • 4. Re: How to open with technical flat calibration in 7.0
                q38 Community Member

                Thank you so much to both Jeff and Noel, you gave me a warm feeling of community in my struggles.

                 

                Noel, of course you are right, I didn't know about that.  That explains why, if I adjusted a DNG flat in 6, then opened it in 7, it would come up with the old sliders in place.  It opens exactly the same in 7 when I choose 2010, right down to the bit level.

                 

                Jeff, I have been learning how to use the DNG Profile Editor you recommended.  When I build a profile  with a linear tone curve and apply it under 2010, all the sliders are left as default, even the "S" tone curve, EXCEPT "Blacks" needs to be set off the default to "0".  When applied under 2012, all the sliders should be set to default EXCEPT Whites to +25 and Blacks to exactly +25.  Even then the highlights are a little different, I haven't measured yet to decide if they are more or less accurate.

                 

                To others in the scientific community:  Even opened linear like this with Adobe RGB (and even extended ProPhoto RGB) there can still be issues with colors going negative, and false colors in highlights as some channels saturate before others.  You can use Adobe DNG Converter to do the dematrixing by choosing Custom > Linear(demosaic) and Uncompressed.  Then "open as" the resulting file in "Photoshop RAW" mode, also known as "RAW RAW" (not camera raw), and choose the appropiate sizing parameters.  With a little trial and error you can get at the raw data right out of the sensors, exactly as it was recorded, except demosaiced.  (If you deselect Linear but keep Uncompressed checked in DNG Converter and open as a single channel, you can actually get under the Bayer array.)  The problem is you are on your own with the tone curve, and some cameras have really weird piecewise linear protocols that can't be described in a continuous equation.  And all metadata is lost in RAW RAW format.  It is probably an impossible dream to wish that Camera Raw could give the straight data off the sensor, calibrated only for gamma 2.2 tone curve.  Such a mode would have no setting for exposure or color temperature as it would always show the full saturation range for each sensor.  Images would look dark and greenish because for a very good reason red and blue sensors are designed to be less sensitive than green.

                • 5. Re: How to open with technical flat calibration in 7.0
                  Noel Carboni Community Member

                  I suspect scientific analysis is exactly in the opposite direction from which Camera Raw is heading, so you may be right about it being an impossible dream.

                   

                  Something to consider might be to look into the various software packages made for astroimaging.  I think you might find that, since astroimagers must do some oddball things (such as stack multiple exposures) the software there may be capable of giving you low level extractions of raw files.  I do know that some astroimagers (myself included) use dSLR gear.  It's been a while since I did it, and the old program I used to use - Images Plus - might be able to see into raw files.  Another one that comes to mind is Maxim DL.

                   

                  -Noel