Smart Eye

Smart Eye is turning edge devices such as Raspberry Pi into an intelligent gateway with deep learning running on it. No internet connection is required, Everything is done locally on the edge device itself. In just three lines of code you can install BerryNet and achieve this deep learning. My goal is to get the camera outside to identify the diverse wild life we have in Prince Edward Island, Canada .The weather on the Island has been nasty so the camera has been delayed going outside.

I used a 16GB SD card and inserted the card to my HP laptop. You can download the Raspberry Pi Imager from here Until the issues are clear with the new Raspberry Image, I'm using the legacy version.

Once the image is complete and place the SD card in the Raspberry PI.Power up the PI and your should see the desktop. Now it's time to install the software

Open the command line and type the following steps. The last step will take a long time to finish. Go for a coffee !

git clone  https://github.com/DT42/BerryNet.git
cd BerryNet
./configure

You want to install the Raspberry Pi camera. This is a perfect tutorial from Raspberry Pi to connect the camera. You can view it here. Now I'm also placing the Pi and the camera into a real nice case from Labists on Amazon.I also brought a tripod that can be attached to the case.

Open the Web Browser in Raspberry Pi and enter this address: http://127.0.0.1/berrynet-dashboard. Select Load Freeboard. Scroll down and select dashboard-tfitedector.json. You should see the camera snapshot and the Inference Result (best guess at what the camera is seeing)