Saturday 15 February 2014

OwlBox or Squirrel Drey

So I've fiddled about with the wireless AV receiver for the OwlBox, and I now have a better idea of whats going on in that box.

I'll have to put the upgrade on hold for this box as a pair of squirrels have made it a home for the time being.

As you can see, the image quality isn't great.
The video is being sent approx 100m via one of these kits from Handycam

This is a 4 channel wireless 2.4 Ghz transmitter / receiver kit.

EzCap analogue to digital USB
This problem with this is that it is interfered with by cordless telephones, wireless routers and microwaves, and also neighbouring systems on the same frequencies.  I think the signal quality degradation is likely to due to this.

Plan B would be to wither use a high gain aerial, or maybe switch to an alternative frequency such as 5.8 Ghz used by professional CCTV applications example here, but this is also not cheap.

Some folk have also tried getting AV  analogue to digital capture devices (EzCap) working on the Raspberry Pi with mixed success, so I think I'll try that next.  See this discussion.  I think this is the best option as I already have network cable to the point of the transmitter for this camera and sending a digital video stream through this is preferable than sending it over the air.

Thursday 13 February 2014

Squirrels (?) in my Owl Box

In my last post I mentioned my owl box.  This was originally designed with Tawny Owls in mind.  Its situated approx 30ft up a conifer tree on the edge of woodland which contains many Tawny owls.  Easy I thought... make the habitat, and they'll move in.  Wrong!

Last year a family of Great Tits moved in and I had to add a cover to the Owl-sized entrance hole so that only my great tits could get in.  We had a competing squirrel who I thought would put off the Great Tits somewhat.  Once these chick fledged, I converted it back to the larger entrance hole, and promptly forgot about it as work stuff too priority.

This was my first camera-equipped nest box and it has a roof-mounted ccd-type analogue camera.  The box is approx 1x1x1 ft in dimensions.  Camera image quality started off good, but has gradually deteriorated.  I'm not sure what went wrong, whether its a combination of the 30ft AV cable run, or the subsequent 150ft wireless transmission over a crowded 2.4Hz frequency, or a dodgy camera, but because of this I have not checked it for several months - plus the PC with the motion capture software suffered a hard drive failure which I have only just got around to fixing.

I turned it all back on this afternoon expecting to get a blue screen, but there is some vestige of signal... and I have a new occupant!  All I can see through the white noise is a fair few small leafy twigs, with "something" moving about   Does not look very much like a small bird nest.

At this stage its not at all clear, but I suspect I may have bad case of squirrel.  I suppose they need somewhere to nest too, so I wont begrudge them their squatters rights.

If I can clear up the video a bit I'll post some if its good enough.

Not a nest of a small bird....

Wednesday 12 February 2014

Bird Box Cameras and the Raspberry Pi

Bird Box Cameras - The journey so far

I'm interested in how technology can be applied so that aspects of the environment around us can be measured, or seen in ways not normally possible.  This project is an alternative take on the basic bird-box camera concept.

I've always had an interest in ornithology, starting from when as a youngster I went out with Dad and watched sea ducks through a telescope approx 1km away in a force 4 gale.  While this might put off some kids, I think this sort of experience drove a need to get things a bit more 'close up'.

As a grown-up I've built some basic bird boxes which have invariably been occupied at one time or another by blue tits & great tits.

Owl box

Last year I started tinkering with camera-enabled bird boxes, and built what I optimistically called my 'owl' nesting box, which was occupied by a family of great tits - nobody told the local Tawny Owls.  This was my first camera-equipped box, which used acolor 540 line CCD camera, from Handykam.  This was linked to a PC video capture card and video captured using iCatcher software from iCode Systems.

Here is some footage of a few early interested visitors: 

In March 2013, we had great fun as a family watching a family of great tits raise a brood. 

The above video shows some male female courting behaviour.  You can tell which one the male is as he has a thicker belly stripe (not evident from this top-down approach!)

9 Great tit eggs
Hungry chicks...

While the video quality started off good, it gradually degraded and is now only works intermittently.  While this is a colour camera, I never got colour video.  I think this is because I used a very long cable run (30m) combined with a 200 ft wireless transmission distance.

Limitations of this approach

1) Relies on radio-frequency transmission, with distance transmission issues
2) Is therefore distance and line-of-site limited
3) Has problems with interference in poor weather - which as I write (Feb 2014) is very poor indeed !
4) Interferes with nearby similar systems - we're a mini-Spring-watch street!
5) No audio

My camera-enabled bird box wishlist

1) Higher resolution, colour video + audio
2) Not reliant on "analogue" radio transmission.
3) Single wire run
4) Compact
5) Preferably internet-protocol (IP) based
6) Low cost

If cost wasn't an issue, there are various ways to achieve this for example, you could theoretically use a dedicated IP-camera & mic option to monitor nesting activity.  

There are many examples out there of internet-enabled wildlife feeds, such as that available for the Peregrines on Norwich Cathedral

Unfortunately, my budget wont stretch to this!
Raspberry-Pi powered birdbox...

I recently discovered the brilliant low-cost computing system that is the Raspberry Pi.  The Pi is a small, bare-board computer that allows you to do a load of cool stuff for minimal cost.  

In this context, by attaching a simple USB webcam to the Pi, and getting creative with some power-over ethernet it is possible to setup a live video stream with one cat5 network cable connected to it.

The following image is a screen capture of a video captured from a high definition USB webcam attached to a Raspberry Pi, powered by power over ethernet, streaming across my home network and motion-captured using iCatcher software.

Bird box camera plus+

I recently found this video from the University of Manchester School of Computer Science which (amongst other things) shows a Raspberry Pi-powered bird box.  

You can read a bit more about this on the BBC Winterwatch blog

The really neat thing they've done here is that you can count nest box activity, which opens up the option to create plots such as these, and lets be honest, we all like a good graph (well, I do anyway!)

The Raspberry Pi can be theoretically put to other uses in this context, for example adding a temperature sensor, and visualising bird activity with the weather.

Anyway, over the next few weeks I'll blog my progress and with luck I'll end up with a high-definition, video streamed video bird box at an affordable cost and have a bit of fun in the process.



Update 10/05/14: My take on this design is described in this post
Update 10/05/14: Description of entrance hole counter in this post