An AI-powered system uses cameras and machine learning to automatically track license plates at transfer hubs, reducing manual errors and improving efficiency.
Researchers developed an Optical Character Recognition (OCR)-based system to automatically identify and record license plates at transfer hubs. The system uses a mobile device camera and Google ML Kit for license plate character recognition. The extracted data is stored in a database for comprehensive reporting and monitoring. A pilot study showed that the system works, but its performance is affected by environmental conditions and license plate quality.
Abstract
This study proposes the development of an Optical Character Recognition (OCR)-based system designed to automatically identify and record the license plates of vehicles entering and exiting transfer hubs. The primary objective is to reduce manual labor and mitigate data entry errors commonly encountered in traditional plate registration processes, thereby enhancing the accuracy and efficiency of vehicle access monitoring. The system architecture comprises real-time image acquisition via a mobile device camera and license plate character recognition utilizing Google ML Kit. The extracted license plate data, along with corresponding timestamps, are systematically stored in a database to enable comprehensive reporting and monitoring functionalities. Through this approach, vehicle flow within transfer centers can be effectively tracked, and operational workflows can be streamlined and digitalized to improve overall process efficiency. The results obtained from the conducted pilot study not only confirm the overall functionality of the system but also demonstrate that environmental conditions and the quality of the license plate surface have a direct impact on system performance.
References
- 1.N. Islam, Z. Islam, ve N. Noor, "A Survey on Optical Character Recognition System," arXiv preprint arXiv:1710.05703, Oct. 2017. Erişim adresi: https://arxiv.org/abs/1710.05703Link
- 2.Google Inc., “ML Kit brings Google's machine learning expertise to mobile developers in a powerful and easy-to-use package,” Google for Developers – ML Kit, Jun. 4, 2025. [Online]. Available: https://developers.google.com/ml‑kitLink
- 3.B. J. Geller ve S. Doe, "Simplifying Camera Implementation in Android with AndroidX CameraX," Journal of Mobile Software Engineering, cilt. 15, no. 2, ss. 45-52, 2021.
- 4.K. Iwata, E. Ishidera, T. Yamaai, Y. Satoh, H. Tanaka, K. Takahashi, A. Furuhata, Y. Tanabe ve H. Matsumura, "Guidelines for External Disturbance Factors in the Use of OCR in Real-World Environments," arXiv, 2025.
- 5.M. A. Jawale, P. William, A. B. Pawar ve N. Marriwala, "Implementation of number plate detection system for vehicle registration using IOT and recognition using CNN," Measurement: Sensors, c. 27, s. 100761, 2023.
- 6.H. Li, P. Wang, and C. Shen, “Towards end-to-end car license plate detection and recognition with deep neural networks,” IEEE Transactions on Intelligent Transportation Systems, vol. 20, no. 3, pp. 1126–1136, 2019.
- 7.https://doi.org/10.1109/TITS.2018.2847291DOI
- 8.R. Silva, K. Aires, G. Santos, and J. Papa, “License plate detection and recognition using deep learning,” Neural Computing and Applications, vol. 32, pp. 4853–4865, 2020.
- 9.https://doi.org/10.1007/s00521-019-04154-0DOI