Accident Prevention: Driver’s Drowsiness DetectionSystem Using AI Techniques

Authors

  • Priya Makarand Shelke
  • Sagar Jadhav
  • Sadanand Nanaware
  • Nikhil Palve
  • Riddhi Mirajkar

Keywords:

Alert, Drowsiness, Drowsy, Eye Aspect Ratio (EAR), MAR, Open CV, Yawn detection

Abstract

One of the main causes of traffic accidents is
driver fatigue and drowsiness. Globally, they
are increasing the number of fatalities and
injuries each year. In this paper, main module
focuses on how to recognize tiredness, which
will assist to decrease accidents and improve
roadsafety. This system gathers photos from a
live webcam feed, applies machine learning to
the image, and determines whether the driver
is sleepy. Many facial expressions and bodily
movements, including yawning and sleepy
eyes, seen as indicators ofsleepiness. The EAR
(Eye Aspect Ratio) measures the ratio of
distances between the horizontal and vertical
eye landmarks to identify sleepiness. The
distance between the lower and upper lips
used to produce a YAWN value for yawn
detection, and the result is compared to a
threshold value. We have used a play sound
library, which will giveappropriate voice alert
messages when the driver is in a drowsy state.
A time limit is applied for driving, if the
driver exceeds the time limit then the also
system gives the alert message. The proposed
system madeto decrease the rate of accidents.

Published

2023-04-26

Issue

Section

Articles