On Thu, 30 Aug 2001 23:50:36 -0600 James Gaa (jamesgaa@telusplanet.net)
wrote:
> Except that when I run the
> cursor over the image and into the area of the scan that is outside of
> the
> image (presumably the unexposed film base), the Nikonscan software shows
> an
> RGB reading of 0,0,6. It seems to me that this part of the scan should be
> 0,0,0.
This is not a fault but is pretty normal for any CCD scanner - the 6B is an
indication of CCD noise. You are quite lucky not to have some Green in
there as well, but Nikons always seem to have a blue noise characteristic.
If the scanner s/w and firmware and calibration are doing their job, you
aren't likely to be able to reduce this value by fiddling with controls.
The usual way to get rid of it is setting the black point so that part of
the tonal range is discarded. Or use multiscanning to reduce the blue by
detecting the random blue pixels on successive passesfrom invariant image
pixels.
At a slightly higher value, you should find the RGB components are all
approximately equal (eg R20 G21 B22). The higher the luminance at which
this happens, the more noise extends into the midrange.
Regards
Tony Sleep
- Online portfolio & exhibit; + film scanner info
& comparisons
