Newsgroups: sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.bluesky.net!news.sprintlink.net!noc.netcom.net!ix.netcom.com!netcom.com!perry
From: perry@netcom.com (Perry West)
Subject: Re: {Q} Best low-medium price CCD camera?
Message-ID: <perryD9u79n.185@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
X-Newsreader: TIN [version 1.2 PL1]
References: <3r50mb$r1@news.cc.geneseo.edu>
Date: Thu, 8 Jun 1995 04:32:11 GMT
Lines: 54
Sender: perry@netcom16.netcom.com

There are at least ten and possibly twenty companies that make cameras 
that meet your price performance requirements.  I don't want to try to 
remember all the names.  You can find the companies in the Photonics 
Buyers Guide.

Here are some things I think you should look for:

1)  Physical size and weight.  The Dage 72 you mentioned is a great 
camera -- but *big* and *heavy*.  Situated atop a microscope, such a 
large camera raises the center of gravity, and contributes to instability 
and image degredation.  There are many fine much smaller and much cheaper 
cameras today.

2)  I assume you will buy a solid-state camera (usually CCD, but there 
are other acronyms that also fit).  The trend is toward smaller image 
sensor sizes.  Smaller sensors mean smaller cameras (usually, but not 
always) and lower costs (less silicon real estate).  As the sensors get 
smaller, the sensing elements also get smaller.  A 2/3 inch sensor is the 
size of the standard vidicon tube, and is fine for almost all 
applications.  The 1/2 inch sensor should be good also.  When you get to 
the 1/3 inch image sensor, the size of the sensing elements is around 5 
to 7 microns.  This is on the order of the blur circle of most high 
quality 35mm camera lenses (e.g., Nikon).  Since most good microscope 
optics are diffraction limited, their blur circle is probably smaller 
than this size.  However, I never use a camera with less than a 1/2 inch 
sensor unless there is absolutely no other option.  Usually, you just get 
empty magnification.

3)  Look for a high image resolution (rows and columns of sensing 
elements).  For standard video in the US you will be limited to between 
480 and 488 active scan lines.  Some cameras offer 492 rows of sensors, 
but the monitors may not display four or so of these rows -- outside of 
the RS170/RS330 spec.  Many microscope applications are using high 
resolution cameras 1,000x1,000 and higher.  But here you are talking 
money above your budget for the camera alone.  The monitor to display the 
greater number of scan lines is also much more expensive.  Great if you 
have the budget.  You should be able to find a good camera with around 
768 columns by 488 rows of sensing elements.  There is one chip made by 
TI and incorporated into cameras by some companies that has around 1,024 
columns x 488 rows.

4)  Watch for specmanship.  Some companies quote sensitivity as the light 
level to give a saturated signal (maximum output).  Others quote the 
light level for the minimum useable (in their opinion) image.  Also, some 
companies are incorporating edge enhancement (video peaking) into their 
cameras.  The effect is to make the image look sharper as if the camera 
had higher resolution.  It also increases the noise somewhat.  The edge 
enhancement is not a problem if all you will be doing is viewing the 
image, but if you anticipate performing any analysis, you might want to 
stay away from this or make sure you can defeat it.

Good luck.
Perry

