Automatic License Plate Recognition: An Efficient and Layout-Independent System Based on the YOLO Detector

Abstract

Automatic License Plate Recognition (ALPR) has been a frequent topic of research due to many practical applications, such as border control and traffic law enforcement. This work presents an efficient, robust and layout-independent ALPR system based on the YOLO object detector that contains a unified approach for license plate detection and layout classification and that leverages post-processing rules in the recognition stage to eliminate a major shortcoming of existing ALPR systems (being layout dependent). We also introduce a publicly available dataset for ALPR, called UFPR-ALPR, that has become very popular, having been downloaded more than 650 times by researchers from 80 different countries over the past two years. The proposed system, which performs in real time even when there are 4 vehicles in the scene, outperformed both previous works and commercial systems on four public datasets widely used in the literature. The entire ALPR system (i.e., the architectures and weights), along with all annotations made by us are publicly available at https://web.inf.ufpr.br/vri/publications/layout-independent-alpr/.

Forensic License Plate Recognition with Compression-Informed Transformers

“…The main focus of ALPR systems is to automatically identify LP strings from (at least partially) controlled acquisition systems of reasonably high quality. Applications include traffic monitoring, automatic toll collection or access control [9]. The systems typically consist of an LP detection step and an LPR step.…”

“…The systems typically consist of an LP detection step and an LPR step. For LP detection, commonly YOLO-based networks are used [9][10][11][12]. LP recognition is generally implemented via convolutional neural networks (CNNs) [9][10][11] or convolutional recurrent neural networks (CRNNs) [12][13][14][15][16].…”

“…For LP detection, commonly YOLO-based networks are used [9][10][11][12]. LP recognition is generally implemented via convolutional neural networks (CNNs) [9][10][11] or convolutional recurrent neural networks (CRNNs) [12][13][14][15][16]. CNN methods classify each character separately.…”

Reconhecimento de Placas Veiculares em Cenários Complexos utilizando o Método do Subespaço Mútuo e Redes Neurais Convolucionais

“…Recentemente, métodos de detecção e classificação de placas baseados em inteligência artificial foram concebidos, tendo resultados destacáveis em acurácia [4], [5], [6], porém, estes métodos demandam mais tempo e poder de hardware do que técnicas tradicionais de visão computacional. Podemos concluir então que embora o reconhecimento automático de placas veiculares seja um desafio bastante explorado nos últimos anos, ainda necessita encontrar um equilíbrio entre baixo custo computacional e alta taxa de precisão.…”