INSTITUTE OF AERONAUTICAL ENGINEERING

(Autonomous)
Dundigal, Hyderabad - 500 043

Academic Year: 2023-24

Department of Data Science
First Review

1
DYNAMIC DRIVER FATIGUE MONITORING THROUGH
VISUAL ANALYSIS
TEAM MEMBERS:

S.NO Name Roll No. Branch & Section

1 G JITHIN 21951A6750 CSE(DS)
2 B JOTHSNA SAI SRI 21951A6753 CSE(DS)
3 R AKSHAYA 21951A6714 CSE(DS)

SUPERVISOR:
Dr. G Ganapathi Rao,Ph.D
Assistant Professor

2
OUTLINE

 Abstract
 Introduction
 Problem Statement
 Proposed solution
 References

3
ABSTRACT

We hear a lot about accidents in our day-to-day life. In India, a person dies in every 4 min
due to road accidents, which is the highest in our world. Though there are several reasons
for accidents to occur, one of the main reasons can be the drowsiness of the driver.
Drowsiness and weariness are among the significant causes of the road accidents.

An alert will be sent to the driver as well as passengers on board if we detect that the
driver is drowsy. This can reduce the probability of an accident to occur and increase
transport safety. First we calculate the eye lid closure, and we take a few frames to
consider and detect the drowsiness which is EAR, and then while he yawns, the values of
the mouth will be calculated in the same procedure as eyes which are MAR.
The message alert will be sent if the values are greater than the threshold value
considered alerting everyone in the vehicle.

4
INTRODUCTION

 In the current state, everyone is busy with their works and they forget to take care of their
health. Health does not stop at being fit. It also includes proper sleep. Sleeping is an
important job every human must do every day to maintain his/her health.

 It is a natural phenomenon that cannot be avoided at all. But, sleeping while driving is not
the proper way. Hundreds of people use the road to travel and it could lead to unexpected
events to occur. Accident due to drowsiness is one of the main factors for the accident on
roads. This could also be caused due to drivers' negligence and carelessness to the road.
These accidents not only affect the driver also the vehicles coming next to it.

 Drowsing while drive doesn't only harm us it also hams our surroundings: our nearby
vehicle is affected by our carelessness. Currently in hurry people avoid taking proper
sleep. This leads to all the problems.

5
PROBLEM STATEMENT

 The solution for driver drowsiness detection using Python is to utilize a combination
of facial landmark detection and eye tracking.

 This solution involves using a live model to detect facial landmarks, such as the
eyes and mouth. Then, by monitoring the position and movement of the eyes, the
system can detect signs of drowsiness, such as drooping eyelids or longer blinks.

 This solution can be implemented using the OpenCV computer vision library in
Python, which provides a range of image processing tools and algorithms.

6
EXISTING SOLUTION

 The solution can also include an alert system to notify the driver when drowsiness is
detected, helping to prevent accidents and improve road safety.

7
PROPOSED SOLUTION
 The camera will be placed in front of the driver to take a snapshot of his face. By
using this camera, we take images of the driver.

 After the face has been recognized, with the use of dlib package, we can plot
landmarks on the detected face. Here, a facial landmark means the location of the
eye and mouth, etc.

 From the eye landmarks EAR value and from the mouth landmarks MAR value are
calculated. Then compare these values with the threshold values of EAR and MAR.

 If the maximum successive frames exceed the thresholds of the EAR and MAR
means the driver is drowsy, then the driver will receive a warning and SMS
messages will be sent to the passengers.

8
PROPOSED SOLUTION
A) Alert System Activation:

 In driver drowsiness detection systems, an alert activation alarm can be triggered

if the Eye aspect ratio (EAR) and Mouth aspect ratio (MAR) cross the threshold
value, indicating that the driver may be drowsy or distracted.

 These ratios change as the eyes close or the mouth opens wider, which can
indicate drowsiness. When the EAR and MAR cross the threshold value, an alert
is activated, which can include a sound, vibration, or visual alert to notify the
driver to take action.

 This approach has been found to be effective in detecting drowsiness and reducing
the risk of accidents due to driver fatigue. By implementing this alert system,
drivers can be alerted before they fall asleep or become distracted, improving road
safety for all.

9
PROPOSED SOLUTION
B)Nearest Hotels:

 In addition to implementing a driver drowsiness detection system, it is also important to

provide drivers with resources for taking rest during night journeys. One way to do this
is to have a webpage that is automatically redirected to a list of nearby hotels and
motels when the driver is detected to be drowsy.

 This can be done by integrating the drowsiness detection system with location-based
services, such as Google Maps.

10
PROPOSED SOLUTION
 This can help the driver to quickly and easily find a place to rest and recharge,
reducing the risk of accidents and improving driver safety. Additionally, this feature
can be combined with other safety features, such as automatic navigation to the
selected hotel or motel, to make the process even more convenient for drivers.

11
PROPOSED SOLUTION

C) SMS Alert System :

 In addition to implementing a driver drowsiness detection system, it can also be

beneficial to have a fail-safe mechanism that alerts the driver's contacts if the driver
becomes drowsy frequently.

 One way to do this is to set up an SMS alert system that sends a text message along
with the driver location to the driver's saved contacts when drowsiness is detected.

 This can be done by integrating the drowsiness detection system with a messaging
service, such as Twilio or Plivo.

12
PROPOSED SOLUTION
 When drowsiness is detected, the system can send a pre-configured message to the
driver's saved contacts, notifying them that the driver may be drowsy and asking them to
check in or take appropriate action.

 This feature can help to ensure that the driver's safety is monitored even if they are not
able to respond to alerts or take action themselves. Additionally, this feature can be
customized to include location information or other relevant details to help the driver's
contacts provide assistance more effectively.

13
REFERENCES

 [1] Vidyagouri H. Hamadr, Umakant P. Kulkami, Detection of Drowsiness

Using Fusional Yawning and Eye Movements. Advances in Computing
Communication, and Control Communications in Copter and Information
Science Volt 3h. pp 583-594 (2017)

 [2] K. & Chidanand Kunar, Brogeshwar Bhowmick, An Application for

Driver Drowsiness Identification based on Pupil Detection using IR
Camera Intelligent Human-Computer Interaction, pp 73-82 (201)

 [3] T.Vesslenyi, S Moca,A. Rus, T. Mitran B. Titaru . Driver drowsiness

detection using ANN image processing (2017)

 [4] Belal Alshaqoq; Abdul Salem Baqubairet; Motame Ou Driver

drowsiness detection system. 2017

14
15