Implementation of YOLOV5 and Tesseract OCR Methods to Detect Vehicle License Plates

Abstract

Campus security and convenience are crucial factors in supporting the advancement of higher education institutions. Conventional security systems that involve manual vehicle inspections through ownership identification often consume time and cause vehicle queues, leading to traffic congestion. To address this issue, this research proposes a solution in the form of an automatic license plate recognition system based on You Only Look Once (YOLO) and character extraction using Tesseract Optical Character Recognition (OCR) technology. This system enables quick and efficient vehicle license plate recognition, optimizing traffic flow, saving time, and enhancing convenience for all vehicle users on campus. The research methodology involves training a YOLO model with a vehicle license plate Dataset to detect and recognize license plates, followed by character extraction to accurately identify plate numbers. The research results show the system's accuracy reaching 70%, indicating its effectiveness in detecting vehicle plates in various situations. It is hoped that this system can be widely implemented on campuses to improve security, convenience, and access efficiency for the entire academic community.