Raspberry Pi Detects License Plates with AI

The Raspberry Pi is a popular choice for makers looking to create home security systems thanks to its ability to use AI. AI systems make it possible to identify potential threats or, in this case, identify the license plate tags on cars that pass by. Maker Yash Indane has created one such license plate identifier, which they call "Platefetcher." Not only does it detect the tag but it also retrieves details about the vehicle registered to that tag and displays the information in a webpage that can be accessed from a mobile device.

The Pi is responsible for handling a few aspects of this project. It reads input from the camera, processes the image with image recognition, and runs the webserver used to host the data it retrieves.

When the Raspberry Pi detects a license plate tag, it parses it through OpenCV and a few other tools to get the exact string for the license plate. This is then sent to a website called Reg Check. This is an open-source API that allows users to search for license plate tags across several European countries.

Image 1 of 2

It doesn’t take too much hardware to recreate this project, though you'll need a Pi powerful enough to handle the AI requirements. Indane is using a Raspberry Pi 4 but you could probably get away with using a Raspberry Pi 3 (or a Raspberry Pi 5 when it’s released). For the camera, any module will do as long as it’s compatible with the Raspberry Pi — Indane is using a USB camera.

The project relies on a few tools including Docker, OpenCV, and AWS Textract. These systems work together to grab the plate data for the Raspberry Pi. Once everything is gathered, the Pi presents the data on a webpage using Flask. You can get a closer look at the source code for this project over at Github.

If you want to see this Raspberry Pi project in action, you can check it out over at the official project page shared to Hackster. Be sure to follow Indane for more cool Pi projects as well as any updates to this one.