P3D Re: Slow Scanning

[Brian Reynolds <reynolds@xxxxxxxxx>]

Michael Kersenbrock wrote:
> > field people thought you couldn't use them outdoors.  These are
> > multiscan devices (R, G and B are exposed seperately through filters
> > in mutilple passes) and you would get all sorts of color fringing due
> 
> 
> Are they all?  I vaguely recall the announcement blurb (in
> Rangefinder magazine) for Leica's digital camera about a year 
> ago having a zillion by 3 CCD that it scanned with.  I assumed that 
> it was doing RGB all at once.  No?
> 

Scanning backs (and scanners) use linear CCDs.  Some of the scanning
backs make three physical passes, some of them have linear CCDs with
three columns of detectors.  Either way the color is not scanned
simutaneously. Think of it this way, while the red column is scanning
a column the green column is one step behind and the blue column is
two steps behind (or whatever order R, G and B are scanned in).  They
can not all be scanning the same coloumn of the image simultaneously.

This is another area where digital can be surpassed by film (depending
on the design and subject matter).  CCDs are not color devices.  The
scanning backs get around this by scanning three times.  If anything
moves there is color fringing.  The area CCDs used in consumer
digicams have a pattern of filters over the sensor so that some of the
detectors are red, some are green and some are blue.  You get color by
interpolating the color values between the sensor sites.  Depending on
the subject this can cause artifacts that I liken to chromatic
aberration.  It isn't actually chromatic aberration since this isn't
caused by the lens not focusing light to one spot, but instead by the
sensor not being sensitive to all wavelengths at all spots.

Astronomical cameras, and some pro cameras (that use area CCDs), avoid
color fringing by taking three (or more) exposures through different
filters.  With the astronomical cameras (and maybe the pro cameras)
you can tailor the exposure times to the detector's sensitivity to a
specific wavelength (or band).  You don't even have to use red, green
and blue filters.  Some amateur astronomers are getting very good
results by taking four exposures similar to CYMK.

-- 
Brian Reynolds                  | "Dee Dee!  Don't touch that button!"
reynolds@xxxxxxxxx              | "Oooh!"
http://www.panix.com/~reynolds  |    -- Dexter and Dee Dee
NAR# 54438                      |       "Dexter's Laboratory"