Actions
Forum Thread Details
  • Replies 24 replies
  • Subscribers 346 subscribers
  • Views 4640 views
  • Users 0 members are here
Related

Choosing a sensor for detecting swimming pool entry

pettitda

I'm looking for ideas on how to sense when a person or animal has entered a swimming pool.  There are several devices on the market that claim to do this, but the reviews are all pretty horrible.  False positives seem to be a big problem with existing products.  I'm thinking that something like an infrared camera would work but it would take considerable effort to set up the detection area (i.e. where the pool is) in the camera's view.  Does anyone have an idea that's not prone to false positives (i.e. leaves falling in the pool, wind blown water, etc)? 

  • I'm not sure it's appropriate to post the link here but...

    There is a project on another website run by another distributor about a wireless water depth detection system.

     

    Essentially is uses an arduino and pressure sensors (BMP280) one measures atmospheric pressure, and the other is housed to measure the pressure exerted by the volume of water. It claims to be accurate to 0.05" on water level! I wondered if something suddenly fell in would the pressure momentarily increase?

     

    My own thought was possibly use an underwater microphone to detect the splash!

     

    Dan.

  • in reply to pettitda

    How about piezo vibration sensors on the rim - they would detect activity on the rim. Or do you only want to know when something goes into the water?

  • in reply to pettitda

    I have an interesting thought.

     

    Why not go with a capacitive sensor, I am thinking about the new TI sense board I have been testing.

     

    You could run a conductive strip around the pool and then connect it as a sensor to the board.  No potential for short here as the sensor is passive.

     

    Anything that approaches to sensor edge will create a detection event.  The sensitivity of the device should cover the full range of your issues.  Plus you can set it up to resolve false detections and identify a full approach and cross sensor profile to look for.

     

    The other solution might be a webcam with change detection software to identify if anything enters the field of view around the pool.

     

    DAB

  • in reply to DAB

    Both ideas are nice DAB, but I see the webcam may have some potential issues like the point of view, availability of a support, different potential detections depending on the field where the event occurs (ok, you say it has just to detect yes/no...) etc. Instead the idea of the capacitive sensor seems better as passive, inexpensive and not-strictly size related. I think it is the worth for a deeper analysis.

     

    Enrico

  • My own thought was possibly use an underwater microphone to detect the splash!

    I was thinking a combination of that and height change due to splash.

     

    Unless the foreign body slowly and gracefully entered the water, it is likely to result in a splash and a wave effect.

     

    mark

  • in reply to pettitda

    I think you're looking at it from the wrong way.   image

     

    Either an electric collar and they'll start barking like mad as the shocks get them.

    It might take a few 'supervised plunges' to discourage them, but eventually they'll decide it's not a good idea.  image

     

    You could also use a collar that when wet, makes a lot of noise ... the dogs could be the detector rather than the pool.image

     

     

    Failing that you need an escape ladder that dumb doggy can use to drag itself out and into humilation.

     

    Mark

  • in reply to balearicdynamics

    Yes, detection is the goal.  Once detected, I may want to either sound an alarm or send a text message or both. 

     

    I think the problem with a camera is determining what is in the pool and what is outside the pool.  The camera only gives a two dimensional representation of the view from it's vantage point.  So, it could be easily fooled into thinking that a person standing between the camera and the pool is actually partially inside the pool.  I suppose if there were two cameras with views at right angles then a quasi 3D representation could be made to reduce this somewhat.

     

    Are you thinking that the capacitive sensor should be inside the pool?  Or on the edge of the pool?  Doesn't the object need to be fairly close to the sensor?  I'm thinking that if the child actually jumped into the pool, they might not come close enough to the sensor to be detected.

  • in reply to mcb1

    My wife has already ruled out shock collars or other similar means of discouragement.  I also want to be able to detect children, especially neighborhood children who's parents might object to them being "collared".  lol

     

    However, we do have an escape ladder than can be used provided the person/animal is capable/intelligent enough to swim to it.

  • in reply to dougw

    I'm primarily concerned with them going into the water.  If they enter the pool via the ladder, it is quite possible to enter without even touching the rim. 

  • When I had a pool, I just put a locking gate around the ladder.