Efficient License Plate Recognition Using Yolov8-Based Detection And CRNN-Based OCR

Abstract

License plate recognition proposes an Efficient Real-Time Vehicle Monitoring and Damage Detection System using YOLOv8 and CRNN models. The system automatically detects vehicle license plates, recognizes plate numbers using OCR, identifies vehicle type and colour and detects visible medium-level dents and scratches through a trained deep learning model. YOLOv8 is used for object detection tasks, including license plate localization, vehicle classification, and damage detection, while CRNN is used for accurate sequence-based text recognition of number plates. The system processes live camera input or uploaded images and displays the results through a web-based interface, enabling intelligent vehicle monitoring in environments such as parking areas, security checkpoints, and institutional campuses. The proposed solution provides an efficient, scalable and real-time approach to automated vehicle analysis using computer vision techniques.