Problem
Research questions and friction points this paper is trying to address.
driver fatigue
railway safety
driver state recognition
alertness monitoring
Graph-Based Recognition of Simulated Train-Driver States From Facial and Upper-Body Keypoints
🤖 AI Summary
This study addresses the limited functionality of conventional systems and the insufficient accuracy of non-contact recognition in railway driver fatigue monitoring by proposing a graph neural network (GNN)-based state classification method using a single RGB camera. By extracting facial and upper-body keypoints, the approach constructs an input representation that fuses facial and skeletal features to accurately classify alert, inattentive, and emergency behavioral states. Additionally, a controlled video dataset encompassing three illumination conditions is established. Experimental results demonstrate that the proposed method achieves 81% accuracy for three-class classification and 99% for binary classification, significantly outperforming single-feature baseline models.
📝 Abstract
Driver fatigue poses a significant challenge to railway safety, with traditional systems like the dead-man switch offering limited and basic alertness checks. This study presents a vision-based monitoring system that relies solely on a single front-facing RGB camera and a graph neural network to classify simulated train-driver states into alert, not-alert, and an emergency class comprising acted emergency-like behaviours. To optimize input representations for the model, an ablation study was performed, comparing three feature configurations: skeletal-only, facial-only, and a combination of both. Experimental results show that combining facial and skeletal features yields the highest accuracy (81%) for the three-class model under the light condition, outperforming models that use only facial or skeletal features. Furthermore, the combination of facial and skeletal features achieves 99% accuracy in the alert/not alert classification in light condition. Additionally, we introduced a controlled RGB video dataset containing alert, not alert, and acted emergency-like behaviours recorded under three illumination conditions. These contributions represent a step toward passive and non-contact train-driver state recognition based on facial and upper-body dynamics.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Graph Neural Network
Driver State Recognition
Keypoint Fusion
RGB Video Dataset
Non-contact Monitoring
🔎 Similar Papers
No similar papers found.
