Egyptian car plate recognition based on YOLOv8, Easy-OCR, and CNN
Amany Sarhan, Rowyda Abdel-Rahem, Bassel Darwish, Arwa Abou-Attia, Ahmed Sneed, Shahd Hatem, A.K. Badran, Mohamed M. Ramadan
Abstract
Abstract This research presents an innovative approach to Egyptian car plate recognition using YOLOv8 and optical character recognition (OCR) technologies. Leveraging the powerful object detection capabilities of YOLOv8, the system efficiently detects car plates within images, videos, or real-time. Subsequently, OCR algorithms are applied to extract alphanumeric characters from the identified plates, facilitating accurate license plate recognition. The integration of YOLOv8 and OCR enhances the system's robustness in varying conditions, contributing to improved performance in real-world scenarios. This study advances the field of automatic license plate recognition, showcasing the potential for practical applications in traffic management, law enforcement, and security systems. A public dataset of Egyptian car plates is used for training and testing the model. Two OCR approaches are used and tested which proved their performance, while CNN-based approach reaches 99.4% accuracy.