h a l f b a k e r yA few slices short of a loaf.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
This is an idea I dreamed up a while ago after seeing an article on the expensive "hawkeye" video system they use at big tennis tournaments to locate where the ball lands.
Instead of using expensive video analysis, use acoustic triangulation to calculate the location of sounds from a tennis court.
Some cheap microphones and a small, cheap, custom DSP device should make this affordable by most clubs.
Placing a handful of microphones (at least 3, perhaps 4) well spaced around a tennis court. Using the differences in timings of sounds arriving at each speaker, it should be possible to build a model of where sounds are being made on the court.
The system can be easily calibrated by issuing a "tick" sound from pre-defined locations on the court after the microphones have been placed.
Accuracy of the system is limited by the speed of sound and the sample rate of the system. Using CD-quality recording (44.1kHz), accuracy would be around 0.7cm (300ms^-1 / 44100 samples sec-1 = 0.0068 meters/sample). Better accuracty could be achieved with a higher sample rate.
Given that there are lots of noises appearing on a tennis court constantly (including echos, just to confuse things), it's going to be difficult to pinpoint which are ball strikes and which are not. It might be possible to build up a "sound image" of the court that could be reviewed at a later date, instead of using it for "live" ball location. Some kind of sound processing might make it possible to identify ball strike sounds, but I haven't thought that bit out. This is the halfbakery, after all.
Acoustic golf ball locator
http://www.iop.org/.../0957-0233/7/12/010
Research done in 1996 [danpat, May 18 2006]
[link]
|
| |
I think the military has something like this to locate snipers. There was talk of it being used to locate the D.C. sniper/s. |
|
| |
So rather than employing expensive video analysis, we're going for expensive acoustic analysis instead? |
|
| |
why not use smell triangulation? Dip the ball in something really stinky. (thinks of other senses) Wait, just use touch! Make the lines and area outside the court pressure-sensitive. |
|
| |
I wonder if there is a type of paint that changes colour with appled pressure, and takes some time ( a couple of seconds) to change back? |
|
| |
Or instead of paint: a Liquid Crystal Device. Try pressing your flat screen display to see what I mean. |
|
| |
It would show footprints as well. |
|
| |
A nice idea, danpat, but in practice probably quite hard to implement properly. As you quite rightly say in your description, there are many other sources of sound in, on, and around a tennis court. Although the sound of a ball hitting could be roughly pinpointed by comparing timings from a video recording to those from a sound track, it could still be easily clouded by background noise. |
|
| |
Furthermore, would the difference in time between the arrival of the sound waves at the various sampling points not be so small over an area the size of a tennis court as to be negligible? Your method might be good for measuring the distance between an observation point and a supersonic jet, but for something so small as a tennis ball I'd make an educated guess at it being somewhat hit-and-miss (pun definitely intended). |
|
| |
Finally, you suggest that the evidence from your system could be reviewed at a later date, but the very purpose of hawkeye is to provide instant analysis. It allows decisions to be made at the time, not after the evidence has been painstakingly reviewed in a lab the week after the event. |
|
| |
A nice idea in principle, but somehow I don't think it would be ideally suited to this particular application. |
|
| |
This is awesome. Kinda like a glorified LCD touch screen. |
|
| |
Good job on the acoustics. [+] |
|
| |