Machine Learning System for Detection of Driver Drowsiness

Vehicle accident is one of the major causes of death in the world of which around 1.3 million people die every year. Of these, at least 20 - 30 percent are caused due to exhaustion of drivers. The involvement of fatigue as a cause is generally grossly underestimated which can be a major reason for contributing in road accidents. Majority of the accidents in recent times are caused because of distraction or drowsiness of drivers. Lack of sleep or distractions like phone calls, talking with the passengers, etc. lead to road accidents. Fatigue combined with bad infrastructure in developing countries like India is a recipe for disaster. To prevent such accidents we propose a system that alerts the driver if the driver gets diverted or feels drowsy. Machine learning trained models for facial landmarks detection are used with the help of Python libraries in the current approach. Need for such driver drowsiness detection systems which is a real time solution for one of the major threats for drivers as well as passengers has also been recently discussed in the Indian Parliament. This whole system will be deployed on a portable hardware which includes a microcontroller, camera for live face image capture and alarms for alerting the driver which can be easily installed in any vehicle for us