Where my nerds at? Electrical engineers and electronics peeps

[Redsand187]

There has got to be some hardcore homeboys here when it comes to building electronics.

I've messed around with Arduino a bit and I want to try a new dumb idea, but can't find the right sensors. I want to detect motion/virtual fence underwater. Super simple to do on dry land, but I can't find any real ways to detect motion or objects underwater and feed it into the controller.

I would imagine a fish finder does what I want, but I can't find any details on how the logic actually works.

Anyone?

 

[chornelium]

There has got to be some hardcore homeboys here when it comes to building electronics.

I've messed around with Arduino a bit and I want to try a new dumb idea, but can't find the right sensors. I want to detect motion/virtual fence underwater. Super simple to do on dry land, but I can't find any real ways to detect motion or objects underwater and feed it into the controller.

I would imagine a fish finder does what I want, but I can't find any details on how the logic actually works.

Anyone?

It's sonar.. at what range are you wanting to detect objects and how large are you talking?

 

[Redsand187]

It's sonar.. at what range are you wanting to detect objects and how large are you talking?

I figure it's sonar, but I can't find shit about that other than talking about big military applications. The range is pretty short, no more than 50ft. Objects fairly decent in size, like the size of a soccer ball or bigger.

 

[Redsand187]

My other thought would be just using a camera with motion detection and zoning. But I don't want to have to deal with processing video just to trigger a relay.

 

[chornelium]

I figure it's sonar, but I can't find shit about that other than talking about big military applications. The range is pretty short, no more than 50ft. Objects fairly decent in size, like the size of a soccer ball or bigger.

My other thought would be just using a camera with motion detection and zoning. But I don't want to have to deal with processing video just to trigger a relay.

Hit up a West Marine and talk to the guys there about the sonar systems and with whatever questions.. there's always a guy they know who specializes in something like that.. the boat scene is weird man, and a great resource.

 

[Ore]

Considering waters index of refraction I don't think optical based motion sensors would be effective. Even a sonar wouldn't be able to exactly give a fence/line of detection.

If you were to use a sonar system I imagine you would need reference objects for the fence i.e. a fixed post is pings off of. Then an algorithm to detect if something moved in front of that post.

I honestly don't know if that is achievable in a hobby electronics Arduino budget but it is an interesting concept.

 

[whatever]

how accurate?

 

[TrenT]

Does a lidar camera work underwater?

 

[Redsand187]

Considering waters index of refraction I don't think optical based motion sensors would be effective. Even a sonar wouldn't be able to exactly give a fence/line of detection.

If you were to use a sonar system I imagine you would need reference objects for the fence i.e. a fixed post is pings off of. Then an algorithm to detect if something moved in front of that post.

I honestly don't know if that is achievable in a hobby electronics Arduino budget but it is an interesting concept.

What I don't get is how fish finders actually work. It should be some type of sonar, but how are they able to tell where a fish is? Like they can show a fish swim by at a specific depth between the surface and the bottom of the lake. In theory, there should be some sort of software solution that could create fencing. If a Chinaman can build a fishfinder for $50 there must be some way to do something similar.

 

[Ore]

What I don't get is how fish finders actually work. It should be some type of sonar, but how are they able to tell where a fish is? Like they can show a fish swim by at a specific depth between the surface and the bottom of the lake. In theory, there should be some sort of software solution that could create fencing. If a Chinaman can build a fishfinder for $50 there must be some way to do something similar.

Fish finders are very general. The big question is how accurate do you want to be?

 

[Redsand187]

Fish finders are very general. The big question is how accurate do you want to be?

Accuracy isn't huge in a sense of exact location or size, just don't want it to miss an elephant in a kiddy pool, but I don't need to see every guppy in lake Michigan. The idea is an input for pool automation to determine if there is anyone in the pool for determining when to run operations.

 

[CUCUY]

Tear into a fish finder and tailor it to what you need.

This would be your best and cheapest option.

 

[Wolf]

Are you looking to place the object IN the water? Is the water relatively clear?

Could you just waterproof a standard camera, send the video feed to a computer (Pi, etc), and then have some image detection software analyze it for movement?

Maybe you could give some details on your use case?

 

[Super Dolemite]

I know a guy with a live feed of his dock in key west, he can watch fish swim by... Why not use underwater video and some video surveillance tech to detect movement and set alarms?

 

[CraneCamsOU]

Accuracy isn't huge in a sense of exact location or size, just don't want it to miss an elephant in a kiddy pool, but I don't need to see every guppy in lake Michigan. The idea is an input for pool automation to determine if there is anyone in the pool for determining when to run operations.

I would think an underwater microphone would be an easier way to go for this application

 

[turbo35]

I went to the store the other day and asked the guy if they had any light sensors I could bypass if I wanted the outlet on full-time. He looked at me and said, "I mean, you could cover the sensor..."

I did a 360 and walked away with my tail between my legs

 

[herpes]

Accuracy isn't huge in a sense of exact location or size, just don't want it to miss an elephant in a kiddy pool, but I don't need to see every guppy in lake Michigan. The idea is an input for pool automation to determine if there is anyone in the pool for determining when to run operations.

Just put a water sensor close enough to the water line where waves from someone swimming will activate it. If the sensor hasn't been triggered in 15 minutes run your "operations"

 

[Dolan]

How big is the pool / the people using it? Might be able to program something to detect relative changes in the water surface level based on the displacement of humans. Obviously the water line will change with evaporation and such, but if the level has increased by 0.0X% relative to 20 minutes ago, someone is probably in it.

 

[herpes]

How big is the pool / the people using it? Might be able to program something to detect relative changes in the water surface level based on the displacement of humans. Obviously the water line will change with evaporation and such, but if the level has increased by 0.0X% relative to 20 minutes ago, someone is probably in it.

Repost

 

[Dolan]

Repost

Where?

Detecting changes in water displacement has nothing to do with detecting waves from motion, if that's what you're getting at.

 

[herpes]

Where?

Detecting changes in water displacement has nothing to do with detecting waves from motion, if that's what you're getting at.

It's literally the exact same thing only my way will work better.

 

[Thumpin'Ram]

Accuracy isn't huge in a sense of exact location or size, just don't want it to miss an elephant in a kiddy pool, but I don't need to see every guppy in lake Michigan. The idea is an input for pool automation to determine if there is anyone in the pool for determining when to run operations.

This pool sensor is actually required by code here. Just Google pool alarm

 

[Dolan]

It's literally the exact same thing only my way will work better.

Weird troll, but alright. Congrats to you.

 

[contactone]

Tear into a fish finder and tailor it to what you need.

This would be your best and cheapest option.

No joke, if you're trying to figure out things on your own this is probably the best way to go about it. At least you get to figure out what part s they but to do the sensing which is probably a bunch of high frequency microphones and an emitter. On the software side you would have to figure out how they translate that sensory input into an image of fish or whatever it is you wanted to detect.

I wish I saw this post before my boss left for the weekend. From what I recall he was trying to map the bottom of some small local lake using his fish finder...I think.

 

[T_Dawgg]

just strap some transmitting electronics to some dolphins chained around the perimeter you want to track