A Semi-Supervised Neural Framework for Real-Time Drowsiness Detection Using Facial Cues

Driver fatigue significantly contributes to road accidents and fatalities, particularly during late-night or early-morning travel due to lack of sleep. Fatigue impairs attention and delays reaction times, increasing accident risk. The American Automobile Association estimates that drowsy driving causes over 320,000 accidents annually, resulting in approximately 6,400 fatal collisions. Building effective drowsiness detection systems has therefore become a paramount need. This paper introduces a novel neural solution for drowsiness detection from the driver’s face using a semi-supervised learning technique. The framework integrates YOLOv8 for accurate face detection and the Swin Transformer to extract subtle fatigue cues using hierarchical attention. An adaptive pseudo-labeling strategy with dynamic confidence thresholding reliably expands the training set, directly improving model robustness. A comprehensive evaluation on the NTHU-DDD, YawDD, and UTA-RLDD datasets demonstrates superior performance, achieving accuracies of 99.99%, 99.34%, and 95.94%, respectively, while maintaining a runtime processing speed of 52.12 frames per second (FPS). The proposed approach outperforms state-of-the-art methods. Ablation studies confirm the effectiveness of both the Swin Transformer backbone and the semi-supervised learning components.