This paper presents IoT-based monitoring system for drowsiness detection for automotive drivers in real-time. The proposed system undergoes three levels of drowsiness detection system to monitor the driver drowsiness and alert him as and when required. The process begins with alcohol detection as a safety precaution, if alcohol is not sensed, then the system proceeds further to detect the face else the engine turns off. Initially, the driver’s face is captured and trained using Haar cascade classifier and AdaBoost algorithm is used to select the meta-data in Haar like features. The proposed system detects only the authorised driver’s face and estimates the eye closure rate, which is captured through the live streaming video from the pi camera. In level 1, if the eye-aspect ratio is below the threshold value, then a sound alerting system is generated. In level 2, if sound alert is prolonged for more than two times, a human voice alerting system is enabled and in the final level, a notification with the GPS location is sent to the driver’s owner or any concerned person. The continuous retrieved data will be stored in the log file. The system uses infrared light to detect driver’s drowsiness at night-time.
Drowsiness Detection for Automotive Drivers in Real-Time
Lect. Notes in Networks, Syst.
Proceedings of Fourth International Conference on Computer and Communication Technologies ; Kapitel : 17 ; 167-189
2023-03-30
23 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Drivers drowsiness detection in embedded system
IEEE | 2007
|Drowsiness Detection for Drivers using IoT
IEEE | 2023
|Drowsiness Detection System in Real Time
Springer Verlag | 2022
|Considerations on Monitoring the Drowsiness of Drivers Through Video Detection and Real-Time Warning
Springer Verlag | 2022
|