Automatic Number Plate Recognition using Machine Learning and Image Processing

Abstract—Over the past few years, the need for intelligent transportation systems has grown tremendously because of the fast growth of urbanization and vehicle density. Effective vehicle monitoring is critical for uses like traffic law enforcement, toll collection, parking management, and access control. Automatic Number Plate Recognition (ANPR) is one of the key technologies supporting these applications and is concerned with detecting and reading vehicle registration numbers from digital video streams or image streams. This work suggests a strong ANPR system based on YOLOv8 for plate localization, EasyOCR for character reading, and SORT tracking algorithm for real-time video processing. Diverse experiments on Indian datasets reveal the high detection precision, robustness in adverse conditions, and real-world applicability of the system in intelligent cities. Index Terms: Automatic Number Plate Recognition, Optical Character Recognition, Post-Processing Optimization, Computer vision, Vehicle Identification.