I am scanning prints with the scanner software SilverFast, which offers the option to scan to a file containing CIELab values. However I need to know what CIE illuminant and observer is assumed when the CIELab values are calculated, and I can't find that information in the documentation. Note that CIELab values are also used when SilverFast does an IT8 calibration of the scanner since it reports a CIE dE* value afterwards. Do you know what CIE illuminant and observer is used for SilverFast CIELab values, or...do you know what illuminant and observer are most common for scanners, or for iT8 color checkers used in scanner calibration? Maybe D50/2 because that is commonly used in graphic arts?
Whenever an illuminant is not specified, assume it's D50 as that is the standard the ICC adapts its profiles to even if the device illuminant differs. For instance a Lab reading in a Photoshop image is relative to D50 whether the image is sRGB or ProPhoto. Scanner profile reference data sets in Lab use D50 even though the scanner's light is rarely close to D50. The target sheets are measured with a spectro.
One interesting thing about Photoshop is that you can read Lab values to fractional precision by setting the info panel to the highest resolution and selecting Lab as the info display space. These values are always shown adapted to D50 white point even if the working space assumes D65. One caution, Photoshop has a slight bug and if you are using 16 bit working spaces and show an image at less than 100% zoom, it shows Lab values erratically varying in the fractional component as if it was converting the RGB values to 8 bits before displaying the values.
I have just been working with my scanner and Silverfast this evening, and has maybe found a break-through, in my attempts to get the color calibration right. I have realized that the provided IT8 reference is measured in CIE D50 Lab values, but the densitometer in Silverfast is showing CIE D65 Lab values.
This has confused me for so long time, where I couldn't get the colors right. Now I have hope, fingers crossed, and looking for a way to convert the D50 values to D65.