Location: Home >> End-to-End Application of Convolutional Neural Networks in License Plate Recognition
License Plate Recognition (LPR) is a critical component of intelligent transportation systems, widely applied in electronic toll collection, traffic violation monitoring, and other fields. Traditional methods rely on manual feature extraction and modular design, leading to insufficient robustness and complex processing pipelines. This paper proposes an end-to-end LPR method based on Convolutional Neural Networks (CNN), integrating license plate detection and character recognition into a unified framework to improve accuracy and computational efficiency. The proposed approach employs an improved detection network to localize license plates, incorporates a Spatial Transformer Network (STN) to rectify tilted plates, and utilizes a Connectionist Temporal Classification (CTC)-based sequence recognition module for character prediction. Experimental results demonstrate that the proposed method achieves high detection and recognition accuracy on public datasets (e.g., CCPD, AOLP), outperforming traditional two-stage approaches and existing deep learning models. This study provides an efficient and robust solution for LPR in complex environments.
[1] Kong H. Research on end-to-end license plate recognition methods in complex scenes [doctoral dissertation]. Nanjing: Nanjing University; 2020.
[2] Chen M, Dai H, Wang S, et al. Implementation of convolutional neural networks in license plate recognition. Wirel Internet Technol. 2024;21(22):62-4.
[3] Qi J, Zhang Z, Fu J, et al. Research on license plate recognition based on convolutional neural network. Technol Innov Product. 2024;45(4):121-4.
[4] Gao Q. Research on license plate recognition in complex environments based on deep learning [master's thesis]. Dalian: Dalian Jiaotong University; 2025.
[5] Cui L. Design and implementation of a vehicle license plate recognition system based on low-resolution images. Electron Technol. 2025;54(6):121-3.
[6] Ji Q, Liu Q, Li M, et al. Research on an improved license plate recognition algorithm under portable charging terminal conditions. J Hebei Univ Water Resour Electr Eng. 2025;35(2):35-41.
[7] Zhu L, Wu X, An C, et al. Design of an automatic license plate recognition system based on the YOLOv8 model. Comput Era. 2025;(8):52-6,60.
[8] Peng H, Ye P. A real-time oversaturation identification method for urban arterials based on license plate recognition data. In: Chi-na Highway and Transportation Society, editor. Proceedings of the 2025 World Transport Convention (WTC2025); 2025; Chengdu, China. Vol. 1, p. 420-6.
Jian Sun. End-to-End Application of Convolutional Neural Networks in License Plate Recognition. OA Journal of Computer Networking, 2025, 4(1), 1-4.
Copyright © 2022 Damray Co., Ltd. Privacy Policy | Terms and Conditions
