:Vehicleoverspeeding is one of the major causes of road accidents, making automated and reliable speed monitoring
systems essential for modern traffic management. This paper presents an AI-based Automatic Number Plate Recognition
(ANPR) system for intelligent vehicle speed detection using a dual-camera approach. The proposed system captures vehicle
images at two fixed checkpoints and calculates the average speed based on the time difference and known distance between the
cameras. Deep learning models are employed to detect vehicles and localize number plates, while an optical character
recognition (OCR) pipeline is used to extract license plate information. To improve recognition accuracy under real-world
conditions, image preprocessing techniques such as contrast enhancement and adaptive thresholding are applied before OCR.
The system supports multiple vehicle categories with predefined speed limits and automatically identifies overspeed violations.
A web-based interface is developed to store records, visualize data, and generate downloadable PDF evidence reports for
detected violations. Experimental evaluation conducted on real vehicle samples demonstrates improved plate recognition
accuracy and reliable speed estimation compared to conventional approaches reported in existing literature. The proposed
system offers a practical and scalable solution for intelligent traffic speed monitoring applications.
[1] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” in
Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2016. [foundation]
[2] Ultralytics, “YOLOv8 Documentation,” 2023. [ultralytics]
[3] Ultralytics, “Ultralytics YOLO GitHub Repository,” 2023.[github]
[4] R. Smith, “An Overview of the Tesseract OCR Engine,” in Proc. Int. Conf. Document Analysis and Recognition
(ICDAR), 2007. [research]
[5] JaidedAI, “EasyOCR: Ready-to-Use OCR with PyTorch,” GitHub Repository, 2020. [github]
[6] A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, “Simple Online and Realtime Tracking,” in Proc. IEEE Int.
Conf. Image Processing (ICIP), 2016.[arxiv]
[7] N. Wojke, A. Bewley, and D. Paulus, “Simple Online and Realtime Tracking with a Deep Association Metric,” in
Proc. IEEE Int. Conf. Image Processing (ICIP), 2017. [arxiv]
[8] OpenALPR, “Open Automatic License Plate Recognition (OpenALPR),” GitHub Repository, 2020. [github]
[9] X. Xu et al., “CCPD: Chinese City Parking Dataset,” GitHub, 2018. [github]
[10] R. Laroca et al., “UFPR-ALPR: A New Benchmark for Automatic License Plate Recognition,” IEEE Access,
2022.[github] Available: https://github.com/raysonlaroca/ufpr-alpr-dataset
[11] AvLab, “AOLP: Application-Oriented License Plate Dataset,” GitHub, 2020. [github]
[12] L. A. Marcomini and A. L. Cunha, “A Comparison between Background Modelling Methods for Vehicle
Segmentation in Highway Traffic Videos,” arXiv preprint, 2018. [arxiv]
[13] Z. Zhang, “A Flexible New Technique for Camera Calibration,” Microsoft Research, Tech. Rep., 2000. [microsoft]
[14] S. Khan and H. Ullah, “A Survey of Advances in Vision-Based Vehicle Re-Identification,” arXiv preprint,2019.[arxiv]
[15] X. Hu, X. Xu, Y. Xiao, H. Chen, S. He, J. Qin and P.-A. Heng, “SINet: A Scale-Insensitive Convolutional Neural
Network for Fast Vehicle Detection,” arXiv preprint, 2018.[arxiv]
[16] GDPR.eu, “General Data Protection Regulation (GDPR) Compliance Guidelines,” 2018. [gdpr]
[17] UK Home Office, “Speedmeter, Traffic Light and Prohibited Lane Enforcement Camera Handbook,” 2023. [assets]
[18] N. Owen, S. Ursachi, and R. Allsop, “The Effectiveness of Average Speed Cameras in Great Britain,” RAC
Foundation, 2016. [racfoundation]
[19] B. Horn and B. Schunck, “Determining Optical Flow,” MIT, 1981. [cmor-faculty]
[20] S. L. Pundlik and A. M. Joshi, “Accurate Vehicle Speed Estimation from Monocular Camera Footage,” ISPRS
Annals, 2020. [isprs-annals]
