Driver Drowsiness Detection System using Machine Learning Approaches

Driver fatigue is related to being a major cause of accidents in many countries of all over the world. These come because of lack of sleep, long working hours or driving, and can cause the driver to be very sleepy hence result in slower response to activities on the road hence increased probabilities of accident. This is an important approach to improving road safety. Scalability of drowsiness identification and the possibility of timely warning is one of the most efficient strategies for avoiding the accidents. With reference to the drowsiness detection techniques, there are physiological measurement, behavioural action and machine learning methods. A solution using ML to achieve the objective of detecting driver drowsiness is presented in this paper. Eye and facial regions are particularized based on a camera, which assesses signs, including the length of an eyelid closure or head tilt. In case of drowsiness is noted, an in-car audible signal is sounded to alert the driver to break the doze. Moreover, safety is improved through the system because it is only possible to receive an email, which has information about the location of the vehicle. While sounding this twin alert minimizes the possibility of an accident, and it also guaranteed assistance to alerted and can get to the scene as earliest as possible.