No abstract

<p><span>Automatic licence plate recognition (LPR) has been a subject of study for the last few decades. Considering the recent advancements in machine learning methods and portable devices, this increasingly attracting researchers’ interest to provide more reliable LPR systems. Several LPR techniques have been reported in the literature in different intelligent transportation applications and surveillance systems, and yet a ropust LPR system remains a challenging research task. Because the performance of current techniques is subject to several factors and local conditions, this paper aims to explore the use of LPR in a specific application; i.e. Automatic plate recognition to monitor the entry and exit of vehicles at the university campus gates. Implementing an auto-gate system is an important application for a smooth control of flowing traffic especially during peak hours. We propose an automated system with the ability to capture, verify and recognize the license plates using image processing-based techniques. The system is aimed to work alongside existing access cards and other gate remote controls. Experimental evaluation of the system reveals a detection accuracy of 91.58%, a successful plate number segmentation rate of 91% and 80% accuracy of plate recognition.</span></p>

Recognizing vehicle plate numbers is a key step towards implementing the legislation on traffic and reducing the number of daily traffic accidents. Although machine learning has advanced considerably, the recognition of license plates remains an obstacle, particularly in countries whose plate numbers are written in different languages or blended with Latin alphabets. This paper introduces a recognition system for Arabic and Latin alphabet license plates using a deep-learning-based approach in conjugation with data collected from two specific countries: Iraq and Malaysia. The system under study is proposed to detect, segment, and recognize vehicle plate numbers. Moreover, Iraqi and Malaysian plates were used to compare these processes. A total of 404 Iraqi images and 681 Malaysian images were tested and used for the proposed techniques. The evaluation took place under various atmospheric environments, including fog, different contrasts, dirt, different colours, and distortion problems. The proposed approach showed an average recognition rate of 85.56% and 88.86% on Iraqi and Malaysian datasets, respectively. Thus, this evidences that the deep-learning-based method outperforms other state-of-the-art methods as it can successfully detect plate numbers regardless of the deterioration level of image quality.