Truck classification on swiss highways using vision and weigh-in-motion systems

Weight-In-Motion (WIM) systems are crucial for detecting vehicle overloads and preventing infrastructure damage. However, their accuracy can be influenced by environmental factors and sensor limitations. This study proposes a vision-based approach for classifying heavy vehicles using the YOLOv5 deep learning model, providing an additional layer to verify and support WIM system outputs. Experimental results demonstrated test accuracy ranging from 96% to 100% for all truck classes. These findings highlight the potential of the proposed approach to improve WIM system reliability.