Improved YOLOv9 for Real-Time Traffic Sign Detection in Intelligent Transportation Systems

Rapid urbanization and population growth have intensified transportation challenges and traffic safety concerns, driving advancements in autonomous driving systems and intelligent transportation systems (ITS). Within this context, accurate and efficient traffic sign detection plays a crucial role in enhancing road safety and optimizing traffic flow. This study presents an improved algorithm based on the YOLOv9 model, tailored for traffic sign detection in ITS applications. The proposed algorithm is evaluated against the baseline YOLOv9 model using a carefully prepared dataset, focusing on key performance metrics. Results demonstrate that the improved YOLOv9 model achieves approximately 3% enhancement in precision, recall, mean average precision (mAP), and F1 score compared to the original model. Furthermore, with a detection time of 8.22 ms, the algorithm is highly suitable for real-time applications. These findings underscore the potential of the enhanced YOLOv9 model in addressing critical ITS challenges and advancing the state-of-the-art in traffic sign detection.