Development and Implementation of a Custom License Plate Detection and Recognition System Using YOLOv10 and Tesseract OCR: A Comprehensive Study in Computer Vision and Optical Character Recognition Technologies

This study presents an automated license plate detection and recognition system, combining YOLOv10 for Realtime object detection and Tesseract OCR for robust text extraction. The methodology involves training a customised YOLOv10 model on annotated vehicle datasets to localize license plates, followed by region-of-interest (ROI) filtering to enhance accuracy. Detected plates are processed with Tesseract OCR to convert visual data into machine-readable text. Evaluated using precision, recall, and inference speed metrics, the system achieves 97 Parsant detection accuracy and real-time performance, demonstrating reliability in automated vehicle identification tasks such as traffic monitoring. This work underscores the synergy of YOLOv10s detection efficiency and Tesseracts OCR capabilities, offering a scalable solution for intelligent transportation systems.