This challenge was designed specifically for the AI Tunisia Hack 2019, which takes place from 20 to 22 September. Welcome to the AI Tunisia Hack participants!
After AI Hack Tunisia, this competition will be re-opened as a Knowledge Challenge to allow others in the Zindi community to learn and test their skills.
Vehicle License plate detection and recognition is a well-known challenge that has been tackled by many computer-vision labs and companies. However, each country has its own specific license plate formats. This challenge is targeting regular Tunisian license plates. The data provided for this challenge is composed of two datasets:
- A set of vehicle images (900 images) taken from the internet and annotated manually. The annotations are the coordinates of the bounding box containing the license plate.
- A set of license plate images (900 images) where the annotations are the text written in the license plate.
This challenge was designed by InstaDeep in Tunisia in partnership with the National Road Safety Observatory of Tunisia, specifically for the AI-Hack-Tunisia 2019 hackathon.
The objective of this challenge is to detect the vehicle’s license plates then recognize the characters in each license plate. The solution will then be used to detect vehicle license plates in traffic cameras.