Your posting didn't provide too much detail about the workflow you are planning. Are you planning to capture, then post-process individual frames after the fact? Also, what sort of resolution/budget trade-offs are you looking at?
If your budget is modest, and critically accurate results are overkill for your intended (educational) purposes, you may be interested in the approach I used for an NI Week demo a couple of years back:
- Camera(s): iBot Firewire webcam and Panasonic PV-DV9000 miniDV camcorder.
At about 7 times the price of the iBot, the optics on the Panasonic camera were obviously superior. Still, the iBot offers manual focus capabilities, and this facilitated adequate image clarity with well-lit subjects at a relatively short distance from the camera (say, between 4 and 8 feet) . (Incidentally, found the results with the iBot to be somewhat better than the results from my cheaper USB webcam.)
- Capture Software: Chris Salzmann's Quicktime VIs (Thanks again Chris!)
If you have the option to consider Macs in your installation, these VIs are certainly worth a look. Easy to integrate, and best of all: FREE! On the tother hand, if you're locked into a "Windows-only" environment, you may need to look for another source (read: $$) for your capture VIs -- although, last I heard, Chris was working on a QT-for -Windows version of his VIs as well...
- Analysis: LabVIEW and Mathematica
Once you have acquired the images into LabVIEW, you will need some analysis tools. If NI's IMAQ tools are a bit pricey, and your school uses Mathematica already, one option is to acquire with LabVIEW and analyze with Mathematica. I used the Mathematica DIP (Digital Image Processing) toolkit, along with "Mathematica Link for LabVIEW" for my demo. And although I have very little background with image processing, I was able to get some interesting results in just a few hours using this combination of software tools. If you are interested in reading more about my modest set-up and demo VIs, you can find an article about it here:
http://www.bettervi.com/mlink/niweek/ML_DIP_Overview.pdf
Of course, the primary economic advantages of this solution presuppose that you have access to the requisite Mathematica licenses -- or, at least, that you would be prepared to invest in Mathematica for this (and perhaps other) projects. If that's the case, the cost of the iBot camera(s), and a site license for "Mathematica Link for LabVIEW" would be negligible compared to other, more elaborate options.
If you have any questions, feel free to get back to me.
Cheers,
Dave Ritter BetterVIEW www.bettervi.com
-----Original Message----- Subject: Firewire video in student labs From: "Bill Gilbert" <[EMAIL PROTECTED]> Date: Fri, 30 Apr 2004 14:14:46 -0500
I've been asked to help in an existing project to use firewire video cams in the student labs, where the idea is to capture streaming video of bouncing balls etc.,then allow the students to make position measurements etc, onscreen in a post-processing application.
If anyone out there is already doing this and has time to share some pointers as to what drivers, cams, etc. has worked for you, we would really appreciate it.
I'm pretty sure I've seen some discussion of this scroll by in the past when I wasn't paying attention to this topic.
Carnegie Mellon Univ has a set of C++ 1394 drivers availiable, anybody using those with dll calls from Labview for camera control?
Thanks,
--BG
Bill Gilbert, EM Tech UMN School of Physics and Astronomy Tel 612 624 4870 Fax 612 624 4578
