Smart Raspberry Pi Bird Feeder Logs Avian Guests

Spring has sprung for makers in the northern hemisphere. The great outdoors are filled with the sounds of chirps, tweets and beautiful songs from migratory birds as they return to their favorite spring nests. Now is a great time for bird tracking projects and no one demonstrates that better than Reddit user MidCitySlim with this Raspberry Pi-powered smart bird feeder.

Using image recognition, this Pi project is programmed to keep track of guests by scanning specifically for birds and storing the images it captures of them. We’ve covered projects in the past that use image recognition for more than just human faces—this Raspberry Pi cat detection system looks for whether or not a cat is holding prey in its mouth before unlocking the pet door. But this is the first one we’ve seen catered primarily to our avian friends.

MidCitySlim’s bird feeder is equipped with an array of sensors for various environmental readings (DHT11 or 22 temperature sensor) that can be accessed in real-time or stored in a log. To trigger the image recognition system, it relies on a ultrasonic sensor. If a bird is detected by the system, it saves the picture. Users can access a custom dashboard to see a live readout of sensor data and look through recently captured images.

Image 1 of 2

The smart bird feeder is built using a Raspberry Pi 3B but there’s no reason you couldn’t use a Raspberry Pi 4 instead. It’s connected to a Raspberry Pi HQ Camera module and relies on a Grove Pi+ HAT for its humidity, temperature, and light sensors.

An ultrasonic sensor is used to detect when guests are present at the bird feeder. This triggers the camera to capture an image. Using Azure, it analyzes the photo to determine if a bird is present. According to MidCitySlim, the image is saved if the probability goes beyond 50% before it's shared using Grafana.

To see this project in action, check out this photo album containing example shots of all of the birds they’ve detected so far. If you want to recreate this Raspberry Pi project yourself, check out the original thread shared to Reddit by MidCity Slim for extra details and an opportunity to inquire for further details.