🚀 go-pugleaf

<1992Sep7.150506.6704@mpifr-bonn.mpg.de>

In <1992Sep7.150506.6704@mpifr-bonn.mpg.de> mlelstv@specklec.mpifr-bonn.mpg.de
(Michael van Elst) writes:

>In <jecome01.715798061@starbase.spd.louisville.edu>
jecome01@rigel.spd.louisville.edu (Eric Comer) writes:
>>      if the lens caps is place on a ccd camera and an image is capture,
>>      why aren't all of the pixel values 0 (black)?  when the image is
>>      normalized it seems like there are faint light diagonal lines
>>      throughtout the whole image.  the image acquired was a gray scale
>>      image the min value was 0 and the max value was 6.  any ideas?

>Well, two things. The CCD doesn't need to give you a value of 0 without
>light. This depends on the offset and gain of the video amplifier. Second,
>your system obviously catches some periodic stray signal. If you know
>your pixel rate then you can translate the number of pixels into the
>offending frequency in the time domain. Most likely this will be a 60Hz
>noise signal.

I tend to agree with Michael, although I'm from the optical side.  So 
count me in as one more vote for the electrical noise, rather than 
optical.

Well, what I meant by optical noise is really the dark current non-
uniformity everybody else mentioned.  What I meant by electrical noise
is the stray signal (such as 60Hz) as quoted above.

The key here is that the levels of signal is rather uniform here, and
form diagonal lines.  For a non-uniformity, usually the first thing you
see is the blemishes, which are like holes in a satuation picture, and
not much in a dark picture.  The second thing you might see, again for
non-uniformity, is the chip granding turns.  These again probably shows
up in lighted pictures better.

Most CCD cameras are not auto-gain-controlled, I believe.  Dark current
noise level (non-uniformity noise) should have been set at a rather low
level, probably 0-3 levels, I would guess.

Am I venturing too far on my guesses?


-- 
Duanfeng He (Jackson)           Duanfeng.He@AtlantaGa.NCR.com