Weather, security and temperature cam

We see a lot of Raspberry Pis being used as security cameras – check out this fine example that we blogged back in 2013 – they’re a cheap and effective solution for people who want to deter burglars and vandals.

This very serious-looking camera housing is only £5.49 on Amazon - click the image to buy.

This very serious-looking fake camera housing is only £5.49 on Amazon – click the image to buy, and then stick a camera board inside.

The good folks at Adafruit had one of those ideas that makes you slap yourself in the forehead for not coming up with it yourself. They’ve made a camera system which can upload images to the cloud, so you can check on it from wherever you are – but it also uploads other sensor data of your choosing (in this example, temperature) and graphs it using matplotlib. A sort of proto-Nest, if you will.

camera_monitor_picam_and_temp_on_pitft v1

We’re using Adafruit’s adafruit.io here: it’s their new Internet of Things API. It’s still in Beta, but pretty solid; we’d be interested to hear how you get on with it.

You can find an exhaustive how-to here. Jeremy Blythe from Adafruit says:

This project uses two Raspberry PIs – a sender and a receiver. The sender has a Raspberry Pi Camera and an MCP9808 temperature sensor to publish data to adafruit.io. The receiver, a dashboard somewhere else in the world, subscribes to this data feed and displays it.

This dashboard Raspberry Pi has a PiTFT and displays the image whenever it’s sent to the feed (every 5 minutes), the current temperature is overlaid on the image using pygame. The final cherry on the cake here is that if you tap the screen you flip to the graph view. This takes the data from the feed using the io-client-python data method, pulls out the last 24 hours and uses matplotlib to draw a graph of temp/time. Of course, you can see the feeds in the adafruit.io online dashboard too!

There’s a lot you can do in terms of feature-creep here; we’re thinking about what other sensors you could usefully add, and what else you might be able to do with a big dataset of images. Go wild – and tell us if you make one yourselves!