A Weather-Adaptive Traffic Light Electrical Detection System in Vietnam: YOLOv5 & PReNet

Accurate traffic light classification in adverse weather conditions is crucial for the safety and efficiency of autonomous driving systems. This research offers a strong method that makes use of YOLOv5 and PreNet to improve traffic signal categorization performance in inclement weather conditions including rain, fog, and snow. For object recognition and initial classification, the fast and accurate YOLOv5 is used, while PreNet, a pre-processing neural network, is used to improve picture quality and reduce the impact of weather-related distortions. By combining the advantages of both networks, the suggested approach improves detection accuracy and dependability in low-light situations. According to experimental data, our system performs much better than traditional approaches, offering a practical alternative for autonomous cars operating in a variety of environmental conditions.