Newsgroups: sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.sprintlink.net!howland.reston.ans.net!swrinde!hookup!nstn.ns.ca!newsflash.concordia.ca!sunqbc.risq.net!hobbit.ireq.hydro.qc.ca!toka.ireq-ccfm.hydro.qc.ca!MEO
From: meo@toka.ireq-ccfm.hydro.qc.ca
Subject: CID/CCD CAMERA QUESTION??
X-Nntp-Posting-Host: toka.ireq-ccfm.hydro.qc.ca
Message-ID: <DBq0IB.5Aw@ireq.hydro.qc.ca>
Sender: news@ireq.hydro.qc.ca (Netnews Admin)
Reply-To: meo@toka.ireq-ccfm.hydro.qc.ca
Organization: Centre Canadien de Fusion Magnetique
Date: Fri, 14 Jul 1995 19:22:59 GMT
Lines: 27

We are presently taking images of a plasma discharges using a CID and CCD BW analog cameras.
We have noticed a distinct behavior of the pixel response when a bright region of the image appears.
I have taken the liberty to test this in the lab using a Lambertian light source. The signal (NTSC)
is digitized in real time using a grabber card. The light source covers a portion of the image.
At a certain power intensity,the pixel resonse seems to be predictable ie: A line scan would show a "background"
at some valueof 12-13 counts (this is an 8-bit digitizer) and a rapid increase at the source.
However as the light intensity is increased to a certain level, the pixel value where the source resides increases 
and the background is"drained". The background values approach 1-5 counts.

I have observed this for both CID and CCD cameras for different positions of the light source in the
image. I have also observed this with a VCR and see the same response (qualitative). I did some research on the technologies of CID and CCD.
The problem may lie in the camera pixel sizes themselves where a strong signal may influence other pixels
(I not talking about blooming). On theory may be that NTSC signal is 1volt peak to peak. Thus a strong signal may imbalance the "black reference".

My question is what causes such a response and what kind of system do we need to avoid this? And in addition
can this be serious in the attempt to calibrate the singal to achieve spatially distributed 
radiated power measurements?

THANK YOU IN ADVANCE FOR YOUR TIME
> ;------------------------------------------------------------------------------
> Fernando Meo
> CCFM
> Tel: (514)-652-1309
> Fax: (514)-652-8625
> E-mail: meo@toka.ireq-ccfm.hydro.qc.ca
> ;------------------------------------------------------------------------------

