Driver Emotion Analytics on In-Vehicle Gesture Control System

Abstract

Automotive safety can be characterized as a mix of design, construction, equipment, and legislation aimed at limiting the negative consequences of motor vehicle collisions. Active Safety Systems help in preventing crashes and accidents by providing advanced warning to the driver. Distracted driving is defined as any activity that draws attention away from the goal of safe driving. Driver drowsiness, weariness, and distraction are common causes of serious accidents around the world. As part of the Active Safety system, the Driver Monitoring System (DMS) is one of the high Automotive Safety Integrity Level (ASIL) which is required for driver aid functionalities, which continuously assists the driver. Constant monitoring of the driver's emotions, head pose estimation and drowsiness will give appropriate warnings to the driver that will aid in preserving road safety. The proposed method will recognize the driver's emotions by detecting the driver's face in the current frame at regular intervals. Deep learning-based algorithms will be used to determine the driver's emotion, drowsiness and head pose and sends alert if necessary. The model accuracy came up to 81.2%. Based on the results obtained, the driver's emotional state, drowsiness and head-pose can be made aware of to the driver if necessary.