PROJECT TITLE :

A New CNN-Based Method for Multi-Directional Car License Plate Detection 

ABSTRACT:

This study presents a novel convolutional neural network (CNN)-based method for high-accuracy real-time car licence plate detection. Many modern methods of detecting a car's licence plate are only good under very specific conditions or under very strong assumptions.

However, they perform poorly if the assessed car licence plate images have a degree of rotation due to the manual capture by traffic police or deviation of the camera. In order to detect car licence plates in both directions, we propose a CNN-based MD-YOLO framework.

Our method, which makes use of precise rotation angle prediction and a fast intersection-over-union evaluation strategy, effectively manages rotational issues in real-time scenarios. Several experiments have been conducted to show that the proposed method outperforms other current state-of-the-art methods in terms of better accuracy and lower computational cost.