Slashdot Mirror
Gunshot Tracking Cameras to be Deployed in LA
apok04 writes "Get out your tinfoil hats (and ski masks). A USC engineer uses his expertise with nerve cells to create a surveillance system that can recognize the sound of a nearby gunshot - and identify the shooter. In a unique pilot program, L.A. and Chicago will deploy test units in high-crime areas. The creator emphasizes that the system cannot recognize voices or words, but his previous research into speech recognition systems suggests otherwise."
The system can then locate, precisely, where the shot was fired, turn a camera to center the shooter in the camera viewfinder and make a 911 call to a central police station.
If the shooter is still there, she deserves to be caught.
According to the article, this device is listening for the entire sound pattern of the gunshot, not just the initial explosion, which makes it much less likely to mistake other loud noises for shooting.
So it may be difficult to fool it unless you can also simulate the whole shooting sequence (think of Matrix's bullet time).
I guess FPS game developers can use one of these to create realistic gunshot sounds.
Rock that crushes, Paper & Scissors that don't matter.
If the surveilance system will determine who fired before it ceases to function due to gunshot damage.
Get out your tinfoil hats
Why?
Doesn't seem like a bad idea to know who's shooting who - don't you think?
In unrelated news, sources report that knive sales have skyrocketed in recent days. No plausible explanation could be found.
Well I guess that just means its time to switch over to my golfball gun or spudgun... Bwa ha ha ha
I don't really know if this is a good or bad thing. I like the idea of having people caught quickly but at the same time I feel that law enforcement agencies would quickly find a way to constantly monitor the cameras, cutting into our privacy even more. Since these cameras are in public it doesn't bother me as much.
Over all I think it's a good idea but it will be exploited so I can't support it fully, even though I'd like to.
An increase in gun-silencers sales has been reported.
I just hope they make it multiplayer and include a deathmatch mode. Also, does the system support skinning?
M
As I recall it turned out that the company doing this was closely affiliated with one of the local politicos and the system was basically bunk. I don't remember how it all played out, but maybe someone else out there does?
If brevity is the soul of wit, then how does one explain Twitter?
Machine sounds are the only ones in SENTRI's vocabulary. It cannot eavesdrop on conversations, the scientist emphasized.
...because we're not done coding that yet, you've got at least another few years.
To fight the war on terror, stop being afraid.
Man on street yells: "Allah Ackbar!!! Allah Ackbar!!!!!". ."
*Directional Finder*: 1) TRIANGULATING... 2) AIMING... 3) FIRING BULLET!
Man on street: "Allah *BAM* Ackkkkkkbahhhhhhhhhhhhhhhhhhhhhrr!!gurgl..lee..l.
SCENE 2
Woman on street whispers to friend: "I hate that dumb idiot Bush"
*Directional Finder*: 1) TRIANGULATING... 2) AIMING... 3) FIRING MIND CONTROL BEAM!
Woman on street whispers to friend: "I.... I... love Bush... and I love Jesus, SUVs, large corporations, and I agree with the righteousness of preemptively saving the rest of the world from themselves and their oil. Let's go shopping."
I remember seeing a system like this years ago. I'm quite sure in the 1980's! Possibly on that Beyond 2000 science show from Australia (we get weird shows here sometimes). I wondered why it was never used it seems like a great invention.
Why so long to get a system like this produced?
Put it in Iraq attached to a machine gun, calibrated to shoot at the sound of an AK-47 not an M16. Since it seems to be able to tune out other explosive noises why not refine it ever further to just a certain gun type?
The device is listening for the entire sound pattern of the gunshot, not just the initial explosion, which makes it much less likely to mistake other loud noises for shooting.
Uh, no, it doesn't. The fact that the guy has worked on different types of signal processing doesn't "suggest" that he builds those capacities into every project he touches.
What I'm listening to now on Pandora...
During the initial studies the camera was placed in front of a TV with Star Wars on it. The sophisticated equipment could still not tell who shot first between Greedo or Han.
In a unique pilot program, L.A. and Chicago will deploy test units in high-crime areas.
Hmmm... Let me guess, the south side of Chicago and Compton?
Rather than looking for pro-active solutions to lowering crime in lower-income neighborhoods, like good education systems, quality health-care, living wages, etc. we continue to see crazy-ass reactive schemes like the above camera system that don't do anything to solve the real problems. In the meantime, as these useless systems become the norm, our society moves closer and closer to the ultimate police heaven, where everyone is monitored every second of every day. When's it gonna end?
Hey, golly-gee-whiz, it sure is a neat technology, Wally.
But like most things of that sort, no one's actually thought about how it actually makes things better, or how it can make things worse. So you catch a few people shooting guns, so what? They end up in jail, their families get torn apart, their chances of actually becoming a productive part of society diminish and they end up back on the street shooting a gun again, which is caught on camera, etc. etc. etc. Wow, crime sure is decreasing now.
It's nice to talk about being tough on crime, but oftentimes what's really needed is not the cracking of a whip, or the monitoring of a camera, but rather a signature on a diploma, or on a paycheck. If you start suspecting everyone as a criminal, then they start seeing themselves as criminals and it becomes a self-fulfilling prophecy. If you first look at people as raw material that can be shaped and molded into something productive, well, you see what I'm getting at.
I'm getting sick of reading about high-tech crime monitoring systems, but it's appearing to be inevitable that we will live with them in our daily lives now and in the near future, so let me practice my indoctrination recitation:
"I for one, welcome our all-seeing camera overlords."
+1 Insightful, -1 Troll. What can I say, I'm an Insightful Troll.
Let's get those combat robots getting sent to Iraq to drive around our cities and automatically counter attack any shooters..... After that all we need is a seriously deranged computer running the whole show and we've got a science fiction movie!
The profile for a voice is much different from that of a gunshot, and creating a multi-purpose system to do this would make both perform much worse.
Also speech recognition knowledge is very different from speaker recognition (one cares about what the person says regardless of how they say it, the other cares about how they say it regardless of what it is). The mathematical models for both are very different.
Also the microphones are likely specialized in the wrong frequency/volume range to be useful for speaker authentication.
1917 ish. Somebody in the Canadian Expeditionary Force to WWI figured out how to accurately direct counter artillery fire using two human ears, a telephone and some trigonometry. The Germans never did figure out why they couldn't fire more than a round or two before they got nailed.
A buddy of mine and I did this in a physics lab. We used an array of 5 condenser mics wired into a PC running LabView, wired out to a laser pointer mounted on some toy motors.
Someone would clap 15-30 feet away, and the computer pointed the laser pointer at their hands. We got the position within a foot or so, even in a echoing cinder block room.
Insights:
- You need at least 4 mics to get an object's position. (There are 4 degrees of freedom, x, y, z, and time) If you only need the angle, then you need 3 (for time, theta, and phi).
- There are some places to shoot where due to the symmetries, it would be hard to compute a position. If the mics are arranged in a plane, then one problem area is straight out from the mic, normal to the plane.
- Another project group in my class developed a computer-controlled ball bearing cannon. I wish we had time to link the projects.
- Thermal variation in the air can disrupt your results.
- If you used well-tuned directional mics, you might be better off. Rather than compute the location based on the path-length of the sound to each mic, you could then find out the incident angle of the sound on each mic, based upon how much the sound level is reduced.
Here's an evaluation. Median location error is about 25 feet. That at least gets it down to two or three houses.
I met the designer of this system some years ago. The original prototype worked using microphones and hard-wired phone connections for each microphone. The signal from each microphone was transmitted using an analog FM carrier system over the phone line designed to trade frequency response for dynamic range. The system had terrible audio frequency response but huge dynamic range, so that pulse events like gunshots come through cleanly without overload. When you have enough dynamic range, gunfire is easy to recognize, because the leading edge of the pulse is so sharp. Few other sounds have that form.
The microphones are up on telephone poles and atop buildings, and they're omnidirectional. So they mostly pick up loud bangs, wind, and aircraft noise. The original pole units were entirely analog, phone line powered, and very dumb. The original central processing system was a PC with some data acquisition cards running LabView. Since then, it's become fancier, with better integration with mapping programs and transmission of gunfire locations to PDA-type devices. But it's not really very complicated.