Résumé:
Traffic accidents caused by a driver falling asleep at the wheel are particularly serious and often fatal, as drowsiness/fatigue seriously impairs the driver's performance, in particular his decision-making ability. Concretely, his reflexes are slowed down or even absent. This is the reason why increased drowsiness causes extremely serious accidents due to loss of control of the vehicle. To avoid this type of accident, it is necessary to detect the unexpected change in the physiological and behavioral state of the driver. In this thesis, we first made a new contribution by evaluating the effect of drowsiness and fatigue on the position of the hands while driving. Then, we proposed a program based on artificial intelligence to detect and calculate the hands on the different positions of the steering wheel automatically and quickly. Finally, we designed and implemented a reliable and mobile driver assistance system to detect driver drowsiness and yawning in real time using a Raspberry Pi microcontroller board and a webcam and some actuators.
