Fatigue driving of drivers will lead to traffic accidents. Effectively monitoring and detecting driver fatigue in real-time is critical for traffic safety enhancement. However, the current approaches usually use a single measure to determine driver fatigue status, which is not robust or comprehensive and may make mistakes in some situations. To this end, a new detection method based on Yolo V5 is proposed to comprehensively judge driver fatigue based on the state of driver eyes and mouth. This method comprehensively considers the state of driver eyes and mouth, which overcomes the limitation of using a single measure. Bio-face and GI4E datasets are used to test the effectiveness of our method. The obtained results show that our method can detect driver fatigue more accurately with a low computation burden. The real-time processing speed of our method is 42 FPS on GTX 1650, which should work in practical applications.
Monitoring and Detection of Driver Fatigue from Monocular Cameras Based on Yolo v5
28.10.2022
1320453 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Lateral Vehicles Detection Using Monocular High Resolution Cameras on TerraMaxTM
British Library Conference Proceedings | 2008
|Monocular 3D Vehicle Detection Using Uncalibrated Traffic Cameras through Homography
ArXiv | 2021
|Speed Estimation of Vehicles Using Monocular Cameras
IEEE | 2023
|