DRIVER DROWSINESS ALERT DETECTION USING MACHINE

LEARNING

1
 OUTLINE
1. Project Title
2. Domain
3. Problem Definition
4. Abstract
5. Scope
6. Existing system
7. Software-hardware requirements
8. Literature Review of Conference/Journal Papers supporting Project
2
idea(minimum 5 papers)
9. Plan of Project Execution
DOMAIN
 Machine Learning:Machine learning is a branch of artificial intelligence (AI) and computer
science which focuses on the use of data and algorithms to imitate the way that humans learn,
gradually improving its accuracy. machine learning (ML) is the area of computational science
that focuses on analyzing and interpreting patterns and structures in data to enable learning,
reasoning, and decision making outside of human interaction.

3
PROBLEM DEFINITION
 In this System, the driver assistance System is presented in order to reduce the number of
accidents caused by driver fatigue and thus improve road safety. This system treats the
automatic detection of face and driver drowsiness based on visual information and artificial
intelligence. We locate, track and analyze both the driver face and eyes to measure by
calculating distance between eye iris and neck angle.

4
ABSTRACT
 we believe that drowsiness can negatively impact people in working and classroom environments as well.
Our solution to this problem is to build a detection system that identifies key attributes of drowsiness and
triggers an alert when someone is drowsy before it is too late. To detect driver drowsiness can be classified
into two categories:
 1) If driver face are detect then this driver is vehicle owner or not.
 2) Behavior-based approaches, and
 3) Eye and neck angle based approaches. In eye and neck angle based approach, the distance of neck angle
and eye iris angle is calculated using Euclidean distance formula.
 Recent studies show that the methods using distance calculation between driver’s eyes are open or closed can
achieve better reliability and accuracy of driver drowsiness detection compared to other methods. The main
idea behind this project is to develop the system which can detect a drowsiness of the driver and issue a
timely warning.

5
OBJECTIVE
• Our proposed method is able to distinguish the simulated drowsy and sleepy states from the
normal state of driving on the low resolution images of faces and eyes observed from an
oblique viewing angle.
• Hence, our system might be able to effectively monitor bus driver’s attention level without
extra requirement for cameras. Our approach could extend the capability and applicability of
existing vision-based techniques for driver fatigue detection.
• To identify the driver is vehicle owner or not.
• To detect driver drowsiness or not using CNN algorithm.

6
 Scope of Project

Proposed system have wide scope some of them are as

follows:
 Avoid Accident.
 Used in School bus or Any vehicles.
 Provide Safety for Passenger.

7
EXISTING SYSTEM
 A driver falls asleep, then the driver loses control over the vehicle, an action which often
results in a crash with either another vehicle or any object. In order to prevent these
devastating accidents, there was the previous approach developed, in this system the state of
drowsiness of the driver was monitored. The following measures were used widely for
monitoring drowsiness:
1. Vehicle-based detection: A number of actions/metrics, including deviations from lane
position, movement of the steering wheel, pressure on the acceleration pedal, etc., are
constantly monitored and any change in these that crosses a specified threshold indicates a
significantly increased probability that the driver is drowsy .
2. Behavioural measures: The behaviour of the driver, including yawning, eye closure, eye
blinking, head pose, etc., was monitored through a camera and the driver was alerted if any
of these drowsiness symptoms are detected.

8
Software Requirements:

• Operating system : 64 bit Windows 10.

• Coding Language : Python

• IDE : Spyder.

• Software Framework : Anaconda Navigator

9
Hardware Requirements:

• System Processors : Core2Duo

• Speed : 2.4 GHz

• Hard Disk : 500 GB

• RAM : 8 GB

10
 LITERATURE REVIEW
SR. Paper Name Author Description
No.
1. Detection of Drowsiness 1.Petchara Inthanon, The purpose of this paper is to develop Algorithm to analyze face structure from
from Facial Images in 2.Surasak Mungsing taking video media and detecting drowsiness. Also, the system is conducted to
Real-Time Video examine Algorithm’s efficiency. The result shows that using facial landmarks can help
Media using Nvidia Jetson to generate eyes and mouth component effectively, which can help to create equations
Nano to analyze drowsiness correctly by using Nvidia Nano Jetson

2. Drowsiness Detection of a Hitendra Garg The developments in technology over the years bring the support to drivers
Driver using Conventional using smart vehicle systems. In past few years, there has been substantial increase in
Computer Vision road accidents in India and worldwide as well. The most significant reasons for the
Applications same are drowsiness and fatigue. Therefore, driver drowsiness and fatigue detection
is major possible area to prevent a large number of sleep induced road accidents
SR. Paper Name Author Description
No.
3. THE SMART AUTOMOBILE Engr. Ghulam Our research aims to serve the purpose of avoiding road accidents caused
(SAM): An Application Based Hyder,Prof. Dr. Bhawani due to driver’s hypo-vigilance, drunk driving or sudden health issues. The
on Drowsiness Detection, Shankar proposed system is an SoC (System on Chip) that can be easily installedin
Alcohol Detection, Vital Sign Chowdhry,Engr. Khuhed the car, it is able to accurately monitor the driver’s state while driving and
Monitoring and Lane based Auto Memon,Engr. Aisha upon observing incapability of driving, the system switches to auto drive
Drive to avoid Accidents Ahmed mode followed
by a safe parking in the left lane

4. Time Domain Parameters as a 1st Venkata In this study, we presented a novel feature extraction strategy based on a
feature for singlechannel EEG- Phanikrishna B,2nd single Hjorth pa_x0002_rameter, and compare its classification capability
based Suchismitha chinara with the existing Power spectral
drowsiness detection method density (PSD) feature. The results show that the proposed H-parameter
features
have higher and stronger performance compared to the PSD features of the
present
work. This field outperforms traditional feature extraction strategies.
SR. Paper Name Author Description
NO.
5. Development of an intelligent Amin Azizi Suhaiman, Several techniques have been studied and analyzed to conclude
drowsiness detection system Zazilah May, Noor A’in the best technique with the highest accuracy to detect driver
for drivers A.Rahman drowsiness. In this work, a real-time system that utilizes
using image processing computerized camera to automatically track and process driver’s
technique eye using Python, dlib and OpenCV is proposed. The eye region
of the driver is measured and calculated continuously to
determine the drowsiness of the driver before triggering an
output alarm to alert the driver.
GAP ANALYSIS
Gap Analysis assesses the difference between the actual and expected performance.
The following points have been analyzed on the basis of which there is a need to elaborate the
existing work:
1. Mostly researcher works on IOT section for detecting drowsiness
2. Mainly, using of two cameras in the system one for monitoring the head movement and the
other one for facial expressions.
3.The other Gap is aging of sensors and all these sensors are attached to the driver's body which
may affect the driver.

14
 Proposed System
• System Architecture –

15
 METHODOLOGY :

The detection of drowsiness can be achieved by analyzing the face area, as signs of drowsiness
are more apparent and easily detectable in this region. From the face area, the location of the
eyes can be determined. Based on eye detection, four types of eyelid movements can be
identified for drowsiness detection: completely open, completely closed, and in transition from
open to closed and vice versa. Figure 1 exemplifies an image used for eyelid movement
detection.

16
 Algorithmic Survey
• CNN :
• Convolutional Neural Networks specialized for applications in image & video
recognition. CNN is mainly used in image analysis tasks like Image recognition,
Object detection & Segmentation.
• There are Four types of layers in Convolutional Neural Networks:
• 1) Convolutional Layer: In a typical neural network each input neuron is connected to
the next hidden layer. In CNN, only a small region of the input layer neurons connect
to the neuron hidden layer.
• 2) Pooling Layer: The pooling layer is used to reduce the dimensionality of the
feature map. There will be multiple activation & pooling layers inside the hidden
layer of the CNN.
• 3) Flatten: - Flattening is converting the data into a 1-dimensional array for inputting
it to the next layer. We flatten the output of the convolutional layers to create a single
long feature vector.
• 4) Fully-Connected layer: Fully Connected Layers form the last few layers in the
network. The input to the fully connected layer is the output from the final Pooling or
Convolutional Layer, which is flattened and then fed into the fully connected layer.

17
 Algorithmic Survey
CNN implementation steps :

Step 1: Convolution Operation(Filter image)

Step 1(b): ReLU Layer
Step 2: Pooling (used max pooling function)
Step 3: Flattening (Covert Matrix into 1D Array)
Step 4: Full Connection.
Step 4(b): Dense()
Step 4(c): Optimizer()
Step 4(d) : compile()

18
19
 Conclusion :

• The face will detect using computer vision and forms

contours around the face. The person is checked with
drowsiness detection then Face detection through the
camera. The set of camera checks for drowsiness
parameters taken by the person. It messages the
concerned person to pick up the person who is being
drowsiness and vehicle owner. The face detection and
driver drowsiness detection through the camera using
harr casscad algorithm and CNN algorithm.

20