Title: Detection and recognition of vehicle licence plates using deep learning in challenging conditions: a systematic review

Authors: Abdul Awal Quraishi; Farid Feyzi; Asadollah Shahbahrami

Addresses: Department of Computer Engineering, Faculty of Engineering, University of Guilan, Rasht, Iran ' Department of Computer Engineering, Faculty of Engineering, University of Guilan, Rasht, Iran ' Department of Computer Engineering, Faculty of Engineering, University of Guilan, Rasht, Iran

Abstract: Automatic licence plate detection and recognition (ALPDR) systems are widely used in various sectors such as traffic control, toll payment, parking systems, border control, and law enforcement. However, these systems face challenges in complex scenarios such as different licence plate formats, poor lighting or occlusion, and deliberate manipulation. To address these challenges, researchers have developed various methods. The first set of challenges involves natural conditions like varying light, snow, rain, fog, and dust. The second set includes environmental factors such as camera angle, occlusion, distortion, contrast issues in images, noise interference, dirt on camera lenses, and camera distance from the scene. The third challenge is related to multinational licence plate variations in terms of formats, colours, sizes, fonts, and characters. Lastly, adversarial attacks pose a threat through rotation, noise addition or distortion to licence plates. This study reviews recent literature on ALPDR systems and proposes guidelines for future research.

Keywords: automatic licence plate detection and recognition; ALPDR; deep learning; DL; character segmentation; character recognition; environmental challenges; multinational challenges; adversarial attacks.

DOI: 10.1504/IJISTA.2024.139736

International Journal of Intelligent Systems Technologies and Applications, 2024 Vol.22 No.2, pp.105 - 150

Received: 20 Jul 2023
Accepted: 02 Nov 2023
Published online: 05 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article