Iranis Large-Scale Character Dataset

Project Card

Category: Computer Vision
Date: January 2021 to September 2021
Affiliation: Research Software Engineer, GSTP, Rasht, Iran
Project URL: [link]
Stack: Python, Keras, Pytorch, YOLO
Related Links: GitHub, Paper, Kaggle

A large-scale dataset of Persian (Farsi) digits and letters with more than 83,000 instances, extracted from Iranian vehicle license plates for Machine Learning and Deep Learning research purposes. The work was funded by Guilan Science and Technology Park and demanded for real-world usage. The dataset is publicly available and the results were published in the paper referenced below:

A. Tourani, S. Soroori, A. Shahbahrami, and A. Akoushideh, "Iranis: A Large-scale Dataset of Iranian Vehicles License Plate Characters," 5th International Conference on Pattern Recognition and Image Analysis (IPRIA), pp. 1-5, Kashan, Iran, 2021, DOI: 10.1109/IPRIA53572.2021.9483461.