Real- Time Driver Drowsiness Detection System using Facial Landmarks

Driver drowsiness is one of the leading causes of accidents and has become a hot research topic. This paper gives an overview of detecting the drowsiness of drivers using behavioral metrics and machine learning approaches. The face imparts a great deal of information (eye blinks, head motions, etc.) which can be utilized to deduce sleepiness. By recognizing the driver's drowsiness and notifying the driver, Computer vision techniques and Image processing technologies can minimize most accidents. This research addresses the issue by identifying key factors such as eye closure, yawning, and head orientation. To determine this, Recurrent Neural Network (RNN) and classifiers were used to extract the facial landmarks, and a 3D locator was used for estimation. In many ways, the culminating results suggest that the real-time approach's performance outperforms the older approach.