Pages: 1 2 3 [4]   Go Down

Author Topic: D800 clipping red channel ? Any possible solutions  (Read 35612 times)

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: D800 clipping red channel ? Any possible solutions
« Reply #60 on: February 12, 2014, 11:21:33 am »

I agree, I'm simply pointing out the vast differences between actual raw data, what may be scene referred data and output referred data, the later more heavily processed data and for some, easier to get at (scene referred more difficult to get to).

This is explained by the ICC in the following paper:

Using ICC profiles with digital camera images:

There is also the case where a camera puts out files containing raw or scene- referred image data. If the raw image data results from capture using a color filter array (e.g. the red, green and blue color values are captured by different pixels), a special camera raw processing application is needed to create a viewable color image. In most cases, these applications (for example Adobe Photoshop camera raw) create standard output referred images, as would the camera.


Here's some further data to chew on from Erick Walowit (this is from the ColorSync list way back in 1998, has anything changed?):

Maybe since 2008, this is changed (Eric's update to the V4 spec, cameras that meet Luther-Ives condition?).

Thanks for the link and info. I don't know if anything has changed or not. Of course the file that I called scene referred is not truly scene referred since no camera meets the Luther-Ives conditions and metameric errors will be introduced. Also flare will reduce the dynamic range. Thus, it is scene referred with limitations.

A simpler way to obtain scene referred data is to change the mode in Photoshop to 32 bit floating point. Starting with ProphotoRGB, the resulting profile listed for the file is "Prophoto RGB (Linear RGB Profile)". Photoshop will not print this file, but it can be imported into Lightroom and printed from there with the usual paper ICC profiles.

Bill
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20614
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D800 clipping red channel ? Any possible solutions
« Reply #61 on: February 12, 2014, 12:15:40 pm »

A simpler way to obtain scene referred data is to change the mode in Photoshop to 32 bit floating point. Starting with ProphotoRGB, the resulting profile listed for the file is "Prophoto RGB (Linear RGB Profile)". Photoshop will not print this file, but it can be imported into Lightroom and printed from there with the usual paper ICC profiles.
By virtue of being 32-bit linear, is it truly scene referred?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: D800 clipping red channel ? Any possible solutions
« Reply #62 on: February 12, 2014, 03:18:38 pm »

By virtue of being 32-bit linear, is it truly scene referred?

Only if the file converted to 32 bit is scene referred. In this case, I used ACR PV2012 with a linear tone curve and rendered into ProphotoRGB and then converted to 32 bit. Converting to untagged linear with ImagesPlus gives a very similar curve, confirming that the rendering is linear. These plots are the bottom patches of the Colorchecker. The ACR conversion is preferable, since it does render into a defined color space, whereas the ImagesPlus is untagged RGB.

Sometimes it is good to have a linear file for photometric measurements and the images are scene referred within the limitations of the CFA sensor.

Bill

Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20614
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D800 clipping red channel ? Any possible solutions
« Reply #63 on: February 12, 2014, 04:19:56 pm »

In this case, I used ACR PV2012 with a linear tone curve and rendered into ProphotoRGB and then converted to 32 bit. Converting to untagged linear with ImagesPlus gives a very similar curve, confirming that the rendering is linear. These plots are the bottom patches of the Colorchecker. The ACR conversion is preferable, since it does render into a defined color space, whereas the ImagesPlus is untagged RGB.

So what's your take based on the above and this definition of scene referred from the ICC?

Quote
scene-referred image data
image data which represents estimates of the colour-space coordinates of the elements of a scene. [ISO 12231]
NOTE 1 Scene-referred image data can be determined from raw DSC image data before colour rendering is performed. Generally, DSCs do not write scene-referred image data in image files, but some may do so in a special mode intended for this purpose. Typically, DSCs write standard output-referred image data where colour rendering has already been performed.
NOTE 2 Scene-referred image data typically represents relative scene colorimetry estimates. Absolute scene colorimetry estimates may be calculated using a scaling factor. The scaling factor can be derived from additional information such as the image OECF, FNumber or ApertureValue, and ExposureTime or ShutterSpeedValue tags.
NOTE 3 Scene-referred image data may contain inaccuracies due to the dynamic range limitations of the capture device, noise from various sources, quantization, optical blurring and flare that are not corrected for, and colour analysis errors due to capture device metamerism. In some cases, these sources of inaccuracy can be significant. ISO 17321-1 specifies a DSC/SMI (DSC Sensitivity Metamerism Index), which can be used to estimate the amount of inaccuracy resulting from capture device metamerism.
NOTE 4 The transformation from raw DSC image data to scene-referred image data depends on the relative adopted whites selected for the scene and the colour space used to encode the image data. If the chosen scene adopted white is inappropriate, additional errors will be introduced into the scene-referred image data. These errors may be correctable if the transformation used to produce the scene-referred image data is known, and the colour encoding used for the incorrect scene referred image data has adequate precision and dynamic range.
NOTE 5 Standard methods for the calculation of scene-referred image data from raw DSC image data will be specified in ISO 17321-2.
NOTE 6 The scene may correspond to an actual view of the natural world, or a computer-generated simulation of such a view. It may also correspond to a modified scene determined by applying modifications to an original scene to produce some different desired scene. Any such scene modifications should leave the image in a scene-referred image state, and should be done in the context of an expected colour rendering transform.

And:

output-referred image data
image data which represents the colour-space coordinates of the elements of an image that has undergone colour rendering appropriate for a specified real or virtual output device and viewing conditions. [ISO 12231]
NOTE 1 The output referred image data is referred to the specified output device and viewing conditions. A single scene can be colour rendered to a variety of output-referred representations depending on the anticipated output viewing conditions, media limitations, and/or artistic intents.
NOTE 2 Output-referred image data may become the starting point for a subsequent reproduction process. For example, sRGB output-referred image data is frequently considered to be the starting point for the colour re-rendering performed by a printer designed to receive sRGB image data.


Apologies if this is OT.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: D800 clipping red channel ? Any possible solutions
« Reply #64 on: February 12, 2014, 06:02:30 pm »

So what's your take based on the above and this definition of scene referred from the ICC?
Apologies if this is OT.

I don't have access to the various ISO standards you reference since they cost around US$150 per copy, but can only rely on the ICC paper regarding obtaining scene referred data via ACR and Photoshop. I did do a brief Google search for ISO 17321 and found two papers that do supply some data of interest.

Link 1

Link 2

The details are quite complicated and beyond my level of expertise, but do indicate that the resulting "scene referred" data that I am obtaining is only an approximation, especially regarding colorimetry. However, the monochrome patches do give linear scene referred responses where the pixel value is proportional to the relative exposure which is derived from the optical densities of the patches converted to reflectance values. The 32 bit files derived from the mode change from 16 to 32 bit floating point are also linear as shown. That is about all I can say. What is your take on the matter?

Regards,

Bill
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20614
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: D800 clipping red channel ? Any possible solutions
« Reply #65 on: February 12, 2014, 06:10:01 pm »

I don't have those ISO papers either but the free PDF from ICC gives some bkgnd on what they consider scene referred (as was Eric's post to the CC list). Perhaps not enough information to extrapolate the full meaning.

My take on the matter based on this is we need to be somewhat careful about what we call scene referred especially if we believes it defines scene colorimtery.

In the end, the process's used either produce results we're happy with or they don't. That's why I take caution using the term 'accurate color'. It does appear some feel the need to jump through extra steps to get a 'rawer' but demosaiced data to begin the edit process, the final results being 'more accurate'. Outside of some (rare) workflows where this may be necessary, I prefer to avoid that and just let the raw converter of my choice produce images as I wish to express them visually.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: D800 clipping red channel ? Any possible solutions
« Reply #66 on: February 12, 2014, 06:20:06 pm »

I don't have those ISO papers either but the free PDF from ICC gives some bkgnd on what they consider scene referred (as was Eric's post to the CC list). Perhaps not enough information to extrapolate the full meaning.

My take on the matter based on this is we need to be somewhat careful about what we call scene referred especially if we believes it defines scene colorimtery.

In the end, the process's used either produce results we're happy with or they don't. That's why I take caution using the term 'accurate color'. It does appear some feel the need to jump through extra steps to get a 'rawer' but demosaiced data to begin the edit process, the final results being 'more accurate'. Outside of some (rare) workflows where this may be necessary, I prefer to avoid that and just let the raw converter of my choice produce images as I wish to express them visually.

I agree completely, but I still think that it is sometimes useful to have quasi scene referred images for some data analysis. Also some postprocessing is best done with linear data, which is why it is best to do as much as possible in ACR/LR. However, linear data can be useful for use in Photoshop where the needed algorithm is not available in ACR.

Regards,

Bill
Logged
Pages: 1 2 3 [4]   Go Up