Abstract
License plate detection and recognition is an urgent topic in traffic regulation due to many practical applications. Globalization and the development of worldwide delivery has contributed to the mixing of vehicles from different countries and regions. This, in turn, imposes additional difficulties for license plate recognition systems due to the difference in plate patterns. The problem of license plate recognition for the CIS countries has not been studied thoroughly. In our study, we propose a license plate detection model based on the YOLOv8 deep convolutional neural network and a multilingual license plate recognition model based on the TrOCR transformer for the CIS countries. The detection model shows a result of 0.983 for mAP@50 metric outperforming baseline in terms of speed and accuracy. The optical character recognition model reaches the best values for the CER metric for the the vast majority of countries of Armenia, Kazakhstan, Ukraine, Moldova, ex-USSR, Kyrgyzstan and also in Europe.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
