Is it right that when a calibrating process is startet the scanned color values are compared to the reference data stored in reference files?
Now what is very interestint to me are the reference values for GS-fields. White is nor white but e.g. 88-0-0 (Lab). That means the reference value is light grey, not white. Theis may be the reason why - when calibrating the scanner - the output color palette is darker than one should expect. Therefore, is it correct that the icc-profile leads to darker scans?
What is the reason for the darker reference values?
Greetings,
Peter
it 8
Dear Peter
yes, you are right concerning scanned color values are compared to color values in the ref file.
However, please not that IT8 calibration is color a calibration, not a brightness calibration process.
What's important therefore, is how the value of any given field (also the fields from the greyscale) compares to the according value noted in the reference file.
Thus it is that the "lightest grey" of the greyscale range does not need to have pure white. Nor does it need to be the same value that's put down in the ref file.
It is relevant to detect the variation, however, between the value given in the file (measured in the lab) and the value the scanner does read from that field. This variation will be used for profiling the scanner.
Or in short: IT8 calibration is about how and not what colors are seen.
Kind regards
Sonny Noack
- tech support, LSI AG -
yes, you are right concerning scanned color values are compared to color values in the ref file.
However, please not that IT8 calibration is color a calibration, not a brightness calibration process.
What's important therefore, is how the value of any given field (also the fields from the greyscale) compares to the according value noted in the reference file.
Thus it is that the "lightest grey" of the greyscale range does not need to have pure white. Nor does it need to be the same value that's put down in the ref file.
It is relevant to detect the variation, however, between the value given in the file (measured in the lab) and the value the scanner does read from that field. This variation will be used for profiling the scanner.
Or in short: IT8 calibration is about how and not what colors are seen.
Kind regards
Sonny Noack
- tech support, LSI AG -
-
pmitterboeck
- SilverFast Beginner
- Posts: 21
- Joined: Sun Oct 03, 2004 10:48 am
Hi!
...returning to this subject after a couple of weeks...
Your response leads to the following question:
If the calibration tells how to see colours and I use this profile when scanning a dia, this should result in image information loss due to the fact that the scan compresses color valuse from the original range of 0 to 100% to say 30 to 100%. Afterwards I have to correct this via histogramm back to 0-100%. Now I have correct colours but aso I have lost 30% of graduation data.
Is this correct???
...returning to this subject after a couple of weeks...
Your response leads to the following question:
If the calibration tells how to see colours and I use this profile when scanning a dia, this should result in image information loss due to the fact that the scan compresses color valuse from the original range of 0 to 100% to say 30 to 100%. Afterwards I have to correct this via histogramm back to 0-100%. Now I have correct colours but aso I have lost 30% of graduation data.
Is this correct???
Dear Peter
why should that be. If your target has a "white" patch which had been read in the lab as 88-0-0 LAB value, and now the scanner reads this patch, say with, e.g. 90-1-0, the generated profile tells SilverFast that image data from the scanner of 90-1-0 should be matched to 88-0-0.
Is does not mean that no "lighter whites" can exist. The ref file just gave a ref value to the "white" patch. There still can be "lighter whites".
Although the leftmost patch of the greyscale is seen in the IT8 norm as the patch with fewest saturation and the rightmost patch with highest saturation, please keep in mind that only printable colours can be achieved on the target. And this is reflected in the IT8 norm (ISO 12641), too.
Kind regards
Sonny Noack
- tech support, LSI AG -
why should that be. If your target has a "white" patch which had been read in the lab as 88-0-0 LAB value, and now the scanner reads this patch, say with, e.g. 90-1-0, the generated profile tells SilverFast that image data from the scanner of 90-1-0 should be matched to 88-0-0.
Is does not mean that no "lighter whites" can exist. The ref file just gave a ref value to the "white" patch. There still can be "lighter whites".
Although the leftmost patch of the greyscale is seen in the IT8 norm as the patch with fewest saturation and the rightmost patch with highest saturation, please keep in mind that only printable colours can be achieved on the target. And this is reflected in the IT8 norm (ISO 12641), too.
Kind regards
Sonny Noack
- tech support, LSI AG -
Return to “Imaging in general”
Who is online
Users browsing this forum: No registered users and 2 guests