Analysis of Research for Fatigue Driving Detection in Complex Scenarios
Sirui Min · Theory and Practice of Science and Technology · 2025
Fatigued driving is a major threat to road traffic safety, and its identification and warning become more challenging in complex environments such as at night, in adverse weather conditions, in mountainous areas, and in urban congestion. This paper examines how complex environmental factors influence fatigue detection and the current research in detection using multimodal perception technology. This includes methods based on physiological signals, as well as techniques for extracting facial landmarks and blink parameters under low-light conditions or facial occlusion. This paper further explores multimodal fusion methods and their application in improving the detection accuracy of environmental factors. Finally, addressing the shortcomings of current research in algorithm robustness, scene generalization capability, and real-road validation, the paper summarizes the challenges faced and outlines future directions aiming to provide references for in-depth research and intelligent prevention and control practices regarding fatigue driving in complex environments.