Traffic Sign Categorization Using YOLO Algorithm: Leveraging Real-Time Object Detection for Improved Road Safety

Traffic signs, and navigation provide drivers with crucial information about speed limits, hazards and are essential for maintaining road safety. Detecting these signs and classifying in real-time plays a vital for autonomous driving systems, as timely information can prevent accidents and ensure compliance with traffic regulations. This paper presents a traffic sign detection system based on the YOLOv8 algorithm, known for its high-speed object detection capabilities, addressing challenges such as occlusion, small object detection, and adverse lighting conditions. The system is trained using a dataset of traffic signs sourced from Roboflow, capturing various types of traffic signs under different environmental conditions, such as lighting, weather, and occlusion scenarios. By using advanced feature extraction techniques, we have visualize the critical features that proposed for the model's classification decisions. The results confirm the efficacy and real-time applications in autonomous vehicles and smart traffic management systems of YOLOv8, indicating the intelligence potential for widespread adoption in the field of vehicles transportation systems.