Bengali license plate recognition system using deep learning
List of Authors
  • Nawal Ayesha Khan , Seam E Nur Sami

Keyword
  • Bangla license plate, deep learning, EasyOCR, recognition, YOLOv8

Abstract
  • Automatic License Plate Recognition (ALPR) systems are a frequent topic of research worldwide and play an essential role in many applications, such as road traffic monitoring and safety, law enforcement, automatic parking system, etc. Traffic violations and crime are prevalent in Bangladesh, so there is an urgent need for a suitable and efficient ALPR system. In this paper, we present a deep-learning approach for license plate detection and recognition for license plates in Bangladesh. The proposed automatic system utilizes the latest object detection algorithm, YOLOv8 and combines it with EasyOCR for the recognition task. The dataset for this paper has been compiled from multiple open-source datasets obtained from Kaggle and photographs taken of vehicles on the road by ourselves. Roboflow framework has been used for preprocessing to build the dataset and mark bounding boxes for the plates. The proposed license plate architecture consists of two stages. In the first stage, YOLOv8 detects or localizes license plates and in the second stage, the EasyOCR package is used to recognize Bengali characters. The YOLOv8 model achieved a mean average precision value of 0.964, while the EasyOCR performed well in the character recognition segment. The experimental results show that the proposed system can accurately localize and detect Bengali characters from the vehicle number plates.

Reference
  • 1. Lubna, N. Mufti and S. A. S. Shah, “Automatic Number Plate Recognition: A Detailed Survey of Relevant Algorithms,” Sensors, vol. 21, 2021.

    2. M. N. I. Suvon, R. Khan and M. Ferdous, “Real Time Bangla Number Plate Recognition using Computer Vision and Convolutional Neural Network”, International Conference on Artificial Intelligence in Engineering and Technology, 2020.

    3. R. Laroca, et al., “A robust real-time automatic license plate recognition based on the YOLO detector,” International Joint Conference On Neural Networks, pp. 1-10, 2018.

    4. H. Sun, M. Fu, A. Abdussalam, Z. Huang, S. Sun, and W. Wang, “License plate detection and recognition based on the YOLO detector and CRNN-12,” International Conference on Signal and Information Processing, Networking and Computers, pp. 66-74, 2019.

    5. S. Abdullah, M. M. Hasan, and S. M. S. Islam, “YOLO-Based Three- Stage Network for Bangla License Plate Recognition in Dhaka Metropolitan City,” International Conference on Bangla Speech and Language Processing, 2018.

    6. N. Saif et al., “Automatic License Plate Recognition System for Bangla License Plates using Convolutional Neural Network,” IEEE Region 10 Conference, pp. 925-930, 2019.

    7. A. Ashrafee, A. M. Khan, M. S. Irbaz and M. A. A. Nasim, “Real-time Bangla License Plate Recognition System for Low Resource Video- based Applications,” Winter Conference on Applications of Computer Vision Workshops, pp. 479-488, 2022.

    8. P. Marzuki et al., “A design of license plate recognition system using convolutional neural network,” International Journal of Electrical and Computer Engineering, vol. 9, 2019.

    9. M. H. Tusar, M. T. Bhuiya, M. S. Hossain, A. Tabassum, and R. Khan, “Real Time Bangla License Plate Recognition with Deep Learning Techniques,” International Conference on Artificial Intelligence in Engineering and Technology, pp. 1-6, 2022.

    10. A. Yeafi (2022). Road Vehicle Images Dataset: Bangladeshi road vehicle images with YOLOv5 annotation. [Dataset] https://www.kaggle.com/datasets/ashfakyeafi/road-vehicle-images- dataset

    11. N. Takbir, S. Afnan and T. Ahmed (2020). Vehicle Detection - Bangladeshi Roads: Pictures with detection and classification bounding boxes for vehicles. [Dataset] https://www.kaggle.com/datasets/nazmultakbir/vehicle-detection- bangladeshi-roads.

    12. S. Chowdhury, S. Chowdhury, J. T. Ifty and R. Khan, “Vehicle Detection and Classification Using Deep Neural Networks,” International Conference on Electrical and Information Technology, pp. 95-100, 2022.

    13. H. H. Shomee and A. Sams, “License Plate Detection and Recognition System for All Types of Bangladeshi Vehicles Using Multi-step Deep Learning Model,” Digital Image Computing: Techniques and Applications, pp. 1-7, 2021.

    14. C. W. Chuang and C. P. Fan, “Deep-Learning Based Joint Iris and Sclera Recognition with YOLO Network for Identity Identification,” Journal of Advances in Information Technology, vol. 12, 2021.

    15. A. Hamdi. Y. K. Chan and V. C. Koo, “A New Image Enhancement and Super Resolution Technique For License Plate Recognition,” Heliyon, vol. 17, 2022.

    16. S. M. Darwish and K. O. Elzoghaly, “An Enhanced Offline Printed Arabic OCR Model Based on Bio-Inspired Fuzzy Classifier,” IEEE Access, vol. 8, pp. 117770-117781, 2020.

    17. N. A. Alam, M. Ahsan, M. A. Based, and J. Haider, “Intelligent System for Vehicles Number Plate Detection and Recognition Using Convolutional Neural Networks,” Technologies, vol. 9, 2021.

    18. S. Siddique et al., “Deep Learning-based Bangla Sign Language Detection with an Edge Device,” Intelligent Systems with Applications, vol. 18, 2023.