Survey on Automatic Drowsiness Detection and Alert System using Deep Learning

Abstract

The majority of deaths in daily life are caused by road traffic crashes. In their daily lives, many people use the highway at all hours of the day and night. Lack of sleep is a problem for long-distance travelers like bus drivers, truck drivers, and taxi drivers. The primary cause of accidents is due to driver fatigue driving. From a recent survey, we came to know that India experiences several million accidents annually. Early detection of drowsiness of driver saves several people’s lives. Therefore, a system employing Python, OpenCV, and Keras is created to prevent these accidents and inform the driver when drowsiness is detected. The webcam images are collected using OpenCV and sent into a transfer learning model like Inception V3, VGG 16, and VGG 19 for learning. This deep neural network works remarkably well for identifying whether a person's eyes are "Open" or "Closed" in an image. The driver will be warned with a loud alarm if the network determines that the eyes have been closed for more than 12–20 consecutive photos. We can use the Python PyGame library for creating video games that are made with handling mouse inputs, playing sound, drawing graphics, etc. If not, it interprets the movement as blinking. Therefore, the Drowsiness Alerting and Detection Technology is a safety system that can stop mishaps brought on by drivers who nod off behind the wheel.