On video in this instance
Forget Hi-8...outdated technology.
Adrian is on to something... www. analysisprogram .com would be perfect for your application. See if he'll send you a test copy...or I could get it and try it...
NOW, the meat of the matter:::::::::
Developing complex video systems for research and high-end sports/entertainment use is my business. A book could literally be written adressing the topics you've brought up. A source for the products you want is B&H in New York. www dot bhphotovideo dot com. They have everything I speak of.
There are three discrete things that you are looking at here
#1 capture
GIGO is the rule, of course.
Tape, and DV specifically, is the only record format to consider today as it is the highest quality, easiest to integrate into a post-production environment, and most tested as well as being the cheapest.
The visual resolution of this and all formats is currently about 720 x 486. While there are some high-end consumer HD cams, their sensors lack some characteristics that are needed in your situation. And their proprietary capture format is a hearache for post and playout.
If I can assume that your budget is not high, I'd suggest a high quality, 3-CCD, DV (tape) camera with a good lens-- a used Canon XL-1 would be best. However, at 1800+, that still may be more than you want to spend. In that case, just about any of the Panasonic consumer cameras with 3-CCDs will be appropriate. Below that, the sony TRV series is excellent.
ALL consumer camcorders record at about 30 frames per second (it is actually much more complicated that that, but for our purposes, let it be). most cams will have the ability to do high shutter speeds, but this will only increase the clarity (less motion blur) of those 1/30 sec time slices. The actual temporal resolution of the recorded material is maxed out at 1/30sec time slices.
If you want 120 frames per second (probably about right for this application, you can hire me and a high-speed film camera, but you'll pay a lot more than the cost of a very nice camcorder.
Wide angle--- most consumer cams' lenses are not wide enough. A wide angle (WA) adaptor is recommended. Spend the extra one on one with a glass lens. Any current cam will have the ability to accept a WA. The lens screw of a camera may have a diameter of 38mm. You'll match that diameter to the range of WA's (and adaptors if your cam uses an odd size) available. Very simple and easy.
#2 Post production.
Once you've got the footage, how to edit and manipulate it (titles, etc)? You cam will have a Firewire (IEEE-1394) I/O port on it. You can buy a hardware/software package for your WINTEL PC for about $300 or less that will have everything you need to do the post AS WELL AS the compression to WEB-compliant and STREAMING video. Finally, it will include software for authoring a DVD (you'll still need a DVD burner, of course)
#3 Output
A DVD is great, probably the best short of giving a DV tape to your customer (as this would not entail an additional codec's distortion-- but DVD's distortion (compression artifacts) are of the same general sort (discrete cosine transform) as the DV codec, so you'll usually not have a great deal of interference distortion created.
NOW---
The problem of a small, fast moving object. An associate of mine developed the Fox "hockey puck" that you may remember from some years past. I think his solution is the inspiration for my solution: rotoscope (think Ralph Bakshi's 'The hobbit'). With a wacom tablet and a suitable editing or compositing program (Premiere or after effects), you could very quickly "paint" over the image of the line in each frame of a few seconds video.
Solution two-- Be extremely careful to have as great a contrast in Hue, Saturation, and Brightness (HSB) between the fiducial (the line) and the background. A bright green line against a dark brown background is MUCH better than that same green line against a blue sky. This will entail selection of the line as well as careful shot framing.
The very best way to do this would be to use multiple cameras (I actually developed a multi-camera system that was used in some major sporting events around the world, that would be perfect for this application, but it's way too expensive, again, to be applicable--- I'm actually going to be doing some consulting work for this company again soon, so i may try to capture myself fly casting with 16 or 32 cameras (think The Matrix meets Derek Brown!). I'll pass along the video when I get it finished.
Cortland (Climax) still makes a black/white line...the Strike Indicator, I think.
Please consider me a resource...
Joe West