Article
Computer Science, Artificial Intelligence
Qi Wang, Xiaocheng Lu, Cong Zhang, Yuan Yuan, Xuelong Li
Summary: This paper constructs a large-scale video-based license plate dataset named LSV-LP, and proposes a new framework called MFLPR-Net to improve the performance of license plate detection and recognition systems.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Software Engineering
The-Anh Pham
Summary: This paper proposes a lightweight and effective deep learning model for license plate detection and recognition. The model differs from traditional methods by discarding max-pooling modules, using alternating convolutional layers and Inception residual networks, and studying different character prediction strategies. Experimental results show a significant improvement in license plate detection and recognition on two public datasets, and the full system can run on low-resource CPU devices at real-time speed.
Article
Computer Science, Artificial Intelligence
Mohammed Salemdeeb, Sarp Erturk
Summary: This study introduces a novel full depth stacked CNN architecture for recognition of handwritten alphanumeric characters and license plate characters, providing low complexity, high accuracy, and full feature extraction. Test results show very low error rates across multiple datasets.
PEERJ COMPUTER SCIENCE
(2021)
Article
Engineering, Electrical & Electronic
Yongjie Zou, Yongjun Zhang, Jun Yan, Xiaoxu Jiang, Tengjie Huang, Haisheng Fan, Zhongwei Cui
Summary: This paper proposes a two-stage license plate recognition algorithm based on YOLOv3 and ILPRNET, which can accurately recognize license plates in complex backgrounds. The algorithm performs well in various complex scenarios, especially achieving recognition accuracy as high as 99.2% in the CCPD series datasets.
SIGNAL IMAGE AND VIDEO PROCESSING
(2022)
Article
Engineering, Civil
Yu Jiang, Feng Jiang, Huiyin Luo, Hongyu Lin, Jian Yao, Jiaxin Liu, Jia Ren
Summary: This paper proposes a real-time and accurate automatic license plate recognition (ALPR) method, which includes a license plate detection network (LPDNet) and character recognition network (CRNet). The method outperforms existing methods in speed and accuracy, especially in complex and challenging environments.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Civil
Sergio M. Silva, Claudio Rosito Jung
Summary: The study presents a complete ALPR system that is able to detect license plates under various conditions and rectify them through Optical Character Recognition methods, achieving satisfactory results.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Civil
Xudong Fan, Wei Zhao
Summary: This paper proposes a robust license plate detection network and a segmentation-free network for accurate license plate recognition and rectification under complex capture scenarios. Experimental results demonstrate the good performance of the system.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Cong Zhang, Qi Wang, Xuelong Li
Summary: License plate detection and recognition has three main issues: poor performance in unconstrained scenarios, limited effectiveness of single-image algorithms, and challenges in complex environments. Our proposed deep learning method successfully addresses these issues and achieves state-of-the-art performance.
Article
Engineering, Civil
Xiao Ke, Ganxiong Zeng, Wenzhong Guo
Summary: This paper presents a two-stage ALPR framework for efficient license plate detection and recognition in various traffic scenarios. The proposed algorithm achieves high accuracy using improved Yolov3-tiny for plate detection and a lightweight recognition network MRNet based on multi-scale features. It also introduces a license plate data augmentation method and provides a challenging dataset for evaluation.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Civil
Yi Wang, Zhen-Peng Bian, Yunhao Zhou, Lap-Pui Chau
Summary: This paper proposes a novel automatic license plate recognition (ALPR) approach called VSNet, which achieves over 50% relative improvement on recognition accuracy based on four key insights including resampling-based cascaded framework, convolutional neural network, vertex information optimization, and weight-sharing character classifier. The proposed VSNet achieves over 99% recognition accuracy on CCPD and AOLP datasets with 149 FPS inference speed, demonstrating outstanding generalization capability on unseen datasets PKUData and CLPD.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Information Systems
Jithmi Shashirangana, Heshan Padmasiri, Dulani Meedeniya, Charith Perera
Summary: This study explores methods and techniques used in automated license plate recognition (ALPR), analyzing related studies in the field and identifying challenges faced by researchers and developers. Future research directions and recommendations are provided to optimize current solutions to operate under extreme conditions.
Article
Computer Science, Information Systems
Tae-Moon Seo, Dong-Joong Kang
Summary: This paper presents a method to improve the accuracy of license plate detection and recognition. The proposed method is more accurate, flexible, and layout-independent. By utilizing lightweight and efficient anchor-free detection networks and attention-based recognition networks with residual deformable blocks, the method has demonstrated its performance in various outdoor conditions.
Article
Computer Science, Information Systems
Qiuying Huang, Zhanchuan Cai, Ting Lan
Summary: The proposed ALPRNet is a neural network designed for detecting and recognizing mixed style license plates, achieving state-of-the-art results without the need for complex recurrent neural network branches.
Article
Chemistry, Multidisciplinary
Musa Al-Yaman, Haneen Alhaj Mustafa, Sara Hassanain, Alaa Abd AlRaheem, Adham Alsharkawi, Majid Al-Taee
Summary: The study proposes an enhanced ALPR system for Jordanian license plates by identifying potential enhancements through ceiling analysis and suggesting improvements, resulting in significant performance improvements without increasing execution time. The experimental evaluation shows superior results compared to equivalent systems, with plate detection accuracy of 94.4%, character segmentation accuracy of 91.9%, and character recognition accuracy of 91.5%.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Hengliang Shi, Dongnan Zhao
Summary: In this paper, an end-to-end deep learning model for license plate location and recognition in natural scenarios was proposed to address the accuracy and speed limitations of traditional license plate recognition methods. The model incorporates an improved channel attention mechanism and location information to enhance feature extraction ability. Parameter reduction and class optimization in the YOLO layer improve efficiency and accuracy in license plate detection. The use of GRU + CTC reduces training time and improves convergence speed and recognition accuracy. Experimental results demonstrate significant improvement over traditional recognition algorithms, with good stability and robustness in complex environments.
Article
Engineering, Electrical & Electronic
Rohit Bhargav, Parag Deshpande
Summary: This paper proposes a method for recognizing License Plate (LP) characters written in English, which is independent of color, scale, and rotation. The method extracts features based on geometrical properties of LP characters and generates a characteristic encoding, enabling the identification of LP characters regardless of their color, scale, and rotational angle. Evaluation using a public dataset resulted in a recognition rate of 98.29% with a processing time of 0.3 ms for a 200 x 100 pixel image. The recognition rate and low processing time compare favorably with other techniques published in the literature, and the proposed method does not impose restrictions on the size, color, or number of characters in LP nor on any specific LP design or region.
IETE JOURNAL OF RESEARCH
(2022)
