h a l f b a k e r yThe embarrassing drunkard uncle of invention.
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.
|
Security systems often involve cameras, each generating a large bandwidth of signal which may be recorded for hours or days on a loop.
Audio, on the other hand, is seldom captured - why? Well, for one thing a microphone covering a large area will not pick up voices from some distance away, over any
background noise. Highly directional microphones are available, but of course they only cover a narrow window.
So, why not have a roughly-spherical (or part-sphere, depending on needs) device, about the size of a football, consisting of a dense array of highly directional microphones, radiating outwards in all directions? The sound from each is recorded, collectively taking up (I would guess) not much more bandwidth than the output of a single camera. In the event of a crime taking place, the sound recorded from the various channels can be played back, until the right channel (right direction) is found. Or, if a camera is mounted nearby, the image would tell you which sound channels are likely to have captured voices.
Sound is, of course, not nearly as useful as video in bringing prosecutions, but it can still be a valuable adjunct to video. With a little audio processing, you would even be able to infer the direction from which sounds have come, even if they were produced outside the camera's field of view.
Not even vaguely related ...
Pop-up_20Sprinkler_...d_20Laptop_20Camera
[normzone, Apr 14 2017]
Related
https://www.ribbonf...e-daredevil-camera/
Using an array of mems mics as a camera for sound [mitxela, Apr 14 2017]
[link]
|
| |
My first thought would be to use a regularly spaced lattice of
omnidirectional mikes, with phase-matched / delay-matched
recording channels, then do what's effectively beam forming
in reverse to get the directionality. |
|
| |
Ok with me. how about "laser listening" where you beam a steerable laser at the moving person to record near person audio? |
|
| |
Laser listening is where a laser is pointed at a wall or window to measure vibrations, computers then translate this to sound. this might possibly work on softer materials. |
|
| |
//regularly spaced lattice of omnidirectional mikes, with phase-matched / delay-matched recording channels// That might work too. Whatever it is needs to be something as cheap(ish) and simple(ish) as a security camera. Or have it combined with a security camera. |
|
| |
They are quite cheap these days, only a few quid each, but a real hassle to solder to. |
|
| |
There's an interesting app note from Invensense where they built a studio-quality microphone by sticking 32 mems microphones onto a single circuit board and summing their output. Pretty cool stuff. |
|
| |
One of the reasons sound isn't commonly recorded for
security purposes could be wiretap laws. |
|
| |
Hmm. That, I had not thought of. You mean that there's a law that prevents you from recording people's voices, even though you can video them? |
|
| |
And, golly gosh, Google tells me you're right. Does that apply in the UK as well, I wonder? |
|
| |