Real-Time Vehicle License Plate Recognition (VLPR) Using Deep CNN

This research introduces a dual-component Vehicle License Plate Recognition (VLPR) system designed to improve the accuracy and efficiency of automated traffic monitoring systems. The first component employs YOLOv8 for real-time detection of vehicle license plates, capitalizing on its advanced capabilities to effectively manage variations in environmental conditions and plate obfuscation. The second component, a custom Convolutional Neural Network (CNN), is optimized for high-precision character recognition from the detected plates. Trained on a dataset of over 33,000 images, the system achieves a detection accuracy of 97.30% and a character recognition accuracy of 98.10%, demonstrating its robustness and effectiveness. This integrated approach not only enhances the reliability of automated traffic monitoring but also holds significant promise for applications requiring high accuracy and real-time data processing across various operational settings.