ðòïåëôù 


  áòèé÷ 


Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 

  óôáôøé 


  ðåòóïîáìøîïå 


  ðòïçòáííù 



ðéûéôå
ðéóøíá












     áòèé÷ :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[filmscanners] Re: Lazy Sensor



I just spent a half an hour writing a reply to this, and Netscape
crashed on my when I went to send it...  Poof! Gone!


One more time (this is going to be much more abridged!)

I think I coined this usage, so I will try to explain...

CCD scanners use a chip which contains one or more lines of sensor
elements, each which responds to light levels.  They are not
manufactured perfectly equally, and over time there response differences.

The Nikon film scanners use one line, and change the light source three
times per scan line (four with IR).  Most others use a triline CCD which
has three lines of CCD elements, each one covered with a filter (RG or B).

To scan the film, the sensor or the film moves a certain distance (2400,
2700, 2820, 4000th of an inch, and the full width is "recorded" at once
by this line of sensor elements.

In order to equalize the response of the individual sensor elements, the
CCD sensors are usually calibrated to a standard white and black point
before each scan.

Sometimes one or more elements are electronically outside of the normal
response curve, usually, it seems slower or less responsive.  When this
happens they respond below the rest and since the element doesn't change
location, a streak or line occurs across the long direction of the scan
or film image. With a Nikon, should this occur (and I haven't heard of
this problem with them) it would be rather obvious, because the one bad
element would be in the same place on all three color scans.

However, in the case of triline CCD sensors, you could have a bad
element at 356 in the blue channel and a bad one at 599 in the red,
etc., so they are less visible.  Sometimes the only way to really see
them is it look at each color channel separately in something like
Photoshop, with the image enlarged to 400% or more.

There are a number of potential causes for a "lazy sensor". The most
obvious is a bad element or series of elements on the CCD, or other
electronics relating to the CCD elements..  But poor calibration
routines could also cause problems, which might be correctable in
software or firmware.  SOme scanners have a calibration chart which the
sensor elements are subjected to to read to determine a standard black
or white and if this has been damaged or become dirty, the CCDS might not
be correctly calibrated.  There could be some dust in the optical path
or the calibration area which is obscuring light from getting to the CCD
sensors properly during calibration or otherwise.

So, that is my meaning for "lazy" sensor elements.  They are not dead,
which is usually really obvious, they just seem to have a response curve
that is out of the norm.

Art


Jack Jansen wrote:

  >>Dave King wrote:
  >>
  >>>Can anything be done about the "lazy sensor" problem?  Is this
something
  >>>that gets more pronounced over time with CCD scanners?
  >>>
  >>>Dave
  >>>
  >
  > Would someone describe the "lazy sensor" problem?
  > Thanks.
  > Jack Jansen
  >
  >
----------------------------------------------------------------------------------------
  > Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
  > or 'unsubscribe filmscanners_digest' (as appropriate) in the message
title or body
  >
  >




----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 
body




 




Copyright © Lexa Software, 1996-2009.