Automatic Number Plate Recognition (ANPR) in Morocco Licensed Vehicles
In Morocco, the number of registered vehicles doubled between 2000 and 2019. In 2019, a few months before lockdowns due to the Coronavirus Pandemic, 8 road fatalities were recorded per 10 000 registered vehicles. This rate is extremely high when compared with other IRTAD countries. The National Road Safety Agency (NARSA) established the road safety strategy 2017-26 with the main target to reduce the number of road deaths by 50% between 2015 and 2026 .
This data-challenge addresses the problem of ANPR in Moroccan licensed vehicles. Based on a small training dataset of 450 labeled car images, the participants have to provide models able to accurately recognize the plate numbers of Morocco licensed vehicles.
The dataset consists of 654 jpg pictures of the front or the back of vehicles showing the license plate. They are of different sizes and are mostly originating from cars. The plate license follows Moroccan standards. The plate strings might contain a series of numbers and latin letters of different length. Because letters in Morocco license plate standard are Arabic letters, we will consider the following transliteration: a ⇔ أ, b ⇔ب, j ⇔ ج (jamaa), d ⇔ د , h ⇔ ه , waw ⇔ و, w ⇔ w (newly licensed cars), p ⇔ ش (police), fx ⇔ ق س (auxiliary forces), far ⇔ق م م (royal army forces), m ⇔المغرب, m ⇔ M.
We provide the plate strings of 450 images (training set). The remaining 204 unlabeled images will be the test set. The participants are asked to provide the plate strings in the test set.
How to Participate ?
- Register on Kaggle
- Register to MoroccoAI Data Challenge: Get Kaggle link
- Join the Competition
- Submit your Predictions
Rules of Participation
- Multiple individuals or entities may collaborate as a team. You may not participate on more than one Team. Each Team member must be a single individual operating a separate Kaggle account
- Team membership may not exceed the Maximum of 3 members
- The participants are encouraged to take further pictures of vehicles in different situations and environments, label them and add them to the training set in order to increase the classification performance of their models
- You agree that upon request from the organizer, you must share your code to allow for reproducibility and checking your result
- Freely & publicly available external data is allowed, including pre-trained models
- Don't cheat!
- Apply yourself!
- Have FUN!
Two prizes will be offered by the ANRT (Agence Nationale de la Réglementation des Télécommunications) and thanks to the Ministry of Digital Transition of Morocco. In addition, Nvidia is offering credits for training courses and certifications by Nvidia Deep Learning Institute (DLI).
The prizes will be awarded on the basis of private leaderboard rank.
* Due to logistics considerations, the participants need to be residents in Morocco to be eligible for the prize
We thank NASSMA-UM6P for providing a part of the dataset and ANRT for sponsoring the prizes and NVIDIA for offering DLI credits.