Driver Drowsiness Detection Using Eye Aspect Ratio (EAR), Mouth Aspect Ratio (MAR), and Driver Distraction Using Head Pose Estimation

Abstract

AbstractMany people are getting injured and losing their lives because of road accidents. One of the main reasons among them is drowsy driving which is the main cause of road accidents and death. In major situations, fatigue is one of the key issues in road accidents. So, we should detect the fatigue in the initial stage, and this has become one of the trending research topics nowadays. Some of the important methods for detection of driver drowsiness are based on behavioral aspects of driver’s face. By using the system, we can detect the face and determine the facial landmarks by which we can compute eye aspect ratio (EAR), mouth aspect ratio (MAR) to detect driver drowsiness based on adaptive threshold, and also using the head pose estimation, which checks the attention of the driver head with respect to the road, whether he is facing the road or not. When the system detects the driver having drowsiness, then it alerts the alarm. Head pose estimation majorly concentrates on three aspects, and they are pitch (used to find direction he is looking left or right), yaw (used to determine looking up or down), and roll (used to determine the rolling of face). In this way, we can detect drowsiness and distraction of driver.KeywordsEye aspect ratio (EAR)Mouth aspect ratio (MAR)Driver drowsinessHead movementYawningFacial landmarkMonitoring system