Road Signage and Road Obstacle Detection Using Deep Learning Method

This study presents a deep learning approach for road signage and road obstacle detection. The purpose of this research was to train a robust and efficient method for detecting road signs and obstacles in real time. This study aims to address the challenges and feasibility of deep learning on road signage and obstacles. A model is trained on YOLOv5 using transfer learning method and the performance of the proposed model was evaluated on a test set. The results showed the YOLOv5 achieved 93.5% mean average precision (mAP). The study concludes that deep learning is a promising method for road signage and road obstacle detection and has potential applications in the field of autonomous vehicles.