Domino60:
That's why the use silencers to make the gun fire almost undetectable.
D.60
You have shown a subsonic hand gun withe the shock wave of a supersonic bullet.
Silencers are of little use with a supersonic rifle round although they may suppress muzzle flash.
Determining direction of a shot in an open space like a desert may be possible but anywhere else requires complex post processing.
How long since JFK was shot ?
I have spent long hours analysing that footage and have come to the conclusion it was suicide , but noone will believe me.
123Splat:
A gun shot within 50 to 75 feet in almost any environ is going to be the loudest noise.
I am pretty sure me mates in the "basement" department will disagree with you there.
Boardburner2:
You have shown a subsonic hand gun withe the shock wave of a supersonic bullet.
Silencers are of little use with a supersonic rifle round although they may suppress muzzle flash.
Determining direction of a shot in an open space like a desert may be possible but anywhere else requires complex post processing.
How long since JFK was shot ?
I have spent long hours analysing that footage and have come to the conclusion it was suicide , but noone will believe me.
I am pretty sure me mates in the "basement" department will disagree with you there.
Wow, I came to the conclusion it was a *Babushka Doryu 2-16 * and the only person who can substantiate that is the CIA operative J Edgar, but once he took of the dress, it blended in with the crowd.
If you have ever noticed, thunder from quite a ways distant is spread out, and has a slow attack and decay time. When it's very close, much sharper.
So a specific type of sudden impulsive sound will have different characteristics, base on distance. How you sort through all of that to distinguish say, a distant gunshot from a nearby door slam, or thunder, or any short impulsive noise is beyond me. Bear in mind that two human ears, with the full auditory capacity of the human brain, can't do this with high reliability. I think it's a nonstarter.
I think it may be possible to detect a rifle bullets source using multiple microphones as it has a distinctive crack.
However (has a bullet been fired or , was that a shot ) is a different question.
jrdoner:
If you have ever noticed, thunder from quite a ways distant is spread out, and has a slow attack and decay time. When it's very close, much sharper.
So a specific type of sudden impulsive sound will have different characteristics, base on distance. How you sort through all of that to distinguish say, a distant gunshot from a nearby door slam, or thunder, or any short impulsive noise is beyond me. Bear in mind that two human ears, with the full auditory capacity of the human brain, can't do this with high reliability. I think it's a nonstarter.
all this is available on the web by those who have done the research.
look at any of the multitudes of people offeing gunshot detectors. in some of the high crime cities, the triangulate the location of the shot(s) can contrary to what the media and politicians want you to believe, there is rarely a man in blue standing there.
I appreciate all the input, opinions and innuendo's. It is my hope we can stick the original request, not tell me what my consequences are. We have a very sophisticated software and false alarms are already handle on our end on a human level. So if you don't have an idea or suggestion on how to detect a gunshot (or a really quick identification of a loud noise such as a gun shot), then please move on. This is not meant to be a smart a** response, just trying stay focused on the idea.
So, does anyone have an idea of a possible design using the arduino to simply detect a gunshot within 50 to 75 feet?
123Splat:
A gun shot within 50 to 75 feet in almost any environ is going to be the loudest noise. Commercial (military) systems look at the profile of the sound received (sharp rise in leading edge, much higher than average ambient level) and look at it coming from multiple sensors, analyze deltas in db and timing.
But, since all you want is an alert, all you have to do is receive sound over a rolling time period, average the intensity to determine a rough ambient level, and alert when the received signal is much, much louder (probably clipping your input network and/or maxing out your ADC output), alert.
So, does anyone have an idea of a possible design using the arduino to simply detect a gunshot within 50 to 75 feet?
An obvious contradiction.
Evidently, you have not gotten the message that "simply detect a gunshot" is not simple. This project is WAY, WAY, WAY above your head, and the chosen hardware.
123Splat:
A gun shot within 50 to 75 feet in almost any environ is going to be the loudest noise. Commercial (military) systems look at the profile of the sound received (sharp rise in leading edge, much higher than average ambient level) and look at it coming from multiple sensors, analyze deltas in db and timing.
You said it yourself , multiple sensors.
A single microphone alone will not always identify it as a shot, multiple sensors can and will also give an approximate position.
The processing i would suggest for this is substantially beyond an arduinos capabilities.
jremington, what makes you think this is WAY WAY WAY above my head? You know nothing about me or my technical and software capabilities...as I asked earlier, if you have an idea, please let me know otherwise move on....geez can I just get an idea if this is possible...if you have other opinions about me, my capabilities or the size of my underwear, keep them to yourself.
You know nothing about me or my technical and software capabilities
.
On the contrary, I know what you have written about the project and that is rather revealing.
From your writings it is absolutely clear that you have done no research on the technical aspects of receiving, digitizing and identifying a particular sound, let alone a sound made by "any gun".
You do not know about the problems of distinguishing a particular sound from irrelevant background noises.
You don't know about the severe limitations of an Arduino and the implications of those limitations for your project.
You have clearly not thought through the consequences of false detections or detection failure.
Your writings suggest that a device that you make might be employed in nightclubs or restaurants, presumably for the purpose of detecting gunshots. I find that frightening, and wish to discourage you.
Again, you know nothing about what I am wanting to do. Please step out of this thread because your assuming is way off base. Give a chance for people that want to help comment.
And actually, I have already designed a circuit that works with our software very well since I started this thread.
Again, your need to bully and comment on threads without helping is worthless to those of us that want to help other users of this forum. Just step out of this one please.
Can you please share that circuit? I'm very interested to see it as this is a common question that comes up regularly on this forum. Perhaps your version of the circuit will become the reference design for hundreds of successful projects.
Graynomad:
They work great but are designed for a gun range where the types of sounds are fairly predictable, IE gunshots and talking. They may not work in say a nightclub with load music, cars backfiring outside etc, but it might be worth looking into the circuitry they use.
Noise cancelling headphones.
Modern BOSE ones are fantastic although expensive, useful in helicopters, nightclubs if you are unfortunate enough to be chucked into one and air travel.
They depend on their action by having a microphone close to the ear to send a cancelling signal to the ear as the sound wave arrives at the ear.
The circuit does not include the Ardunio and requires our software to perform in the environment. We are going to include it in our patent pending so I cannot share at this time. But once the filing is completed, I will share.
Modern BOSE ones are fantastic although expensive, useful in helicopters, nightclubs if you are unfortunate enough to be chucked into one and air travel.
They depend on their action by having a microphone close to the ear to send a cancelling signal to the ear as the sound wave arrives at the ear.
They also depend on the sound having a repeating pattern. That's how they can identify some frequencies (noise) for more attenuation than others (conversation.) I suspect that the normal Bose headphones would not cancel the sound of a gunshot very well. On page 1 of this thread, someone suggested that there may be headphones designed for this specific noise cancellation.
