ANPR v2
Building upon my initial project, I further improved the system by integrating the YOLOv8 model to enhance detection accuracy and performance. Unlike the previous version, which primarily relied on OpenCV and EasyOCR for image-based detection, this new version supports real-time license plate recognition from video input.
Development Journey
As I moved forward with the enhanced version, I delved deeper into researching different machine learning models to find the most suitable one for license plate detection. I explored various architectures, comparing their performance in terms of accuracy, speed, and efficiency when processing license plate datasets. After careful evaluation, I decided to implement YOLOv8, which provided superior results in real-time detection from video input. This journey has been a transformative learning experience, pushing me to develop skills in computer vision, model optimization, and deep learning implementation.