Driver Drowsiness Detection System

10 V May 2022
https://doi.org/10.22214/ijraset.2022.42109
International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 10 Issue V May 2022- Available at www.ijraset.com
Driver Drowsiness Detection System

Manjunath S1, Banashree P2, Shreya M3, Sneha Manjunath Hegde4, Nischal H P5
1
Assistant Professor, Department of Information Science and Engineering, Global Academy of Technology, Bangalore
2, 3, 4, 5
Student, Department of Information Science and Engineering, Global Academy of Technology, Bangalore
Abstract: Recently, in addition to autonomous vehicle technology research and development, machine learning methods have
been used to predict a driver's condition and emotions in order to provide information that will improve road safety. A driver's
condition can be estimated not only by basic characteristics such as gender, age, and driving experience, but also by a driver's
facial expressions, bio-signals, and driving behaviours. Recent developments in video processing using machine learning have
enabled images obtained from cameras to be analysed with high accuracy. Therefore, based on the relationship between facial
features and a driver’s drowsy state, variables that reflect facial features have been established. In this paper, we proposed a
method for extracting detailed features of the eyes, the mouth, and positions of the head using OpenCV and Dlib library in order
to estimate a driver’s level of drowsiness.
Keywords: Drowsiness, OpenCV, Dlib, facial features, video processing
I. INTRODUCTION
One of the main reasons for untimely deaths today is road accidents. In the current times, we can see drastic changes in how humans
manage their time. The natural sleep cycle of human beings has therefore been disturbed. Due to a lack of sleep and irregular sleep
cycles, humans tend to feel drowsy at any time of the day. With these poor work-life timings, people can find it difficult to carry out
the activities like driving which requires a healthy and properly functioning state of mind and body. Drowsiness is one of the major
causes of road accidents in today’s time. According to the Central Road Research Institute (CRRI), tired drivers who drowse off
while driving are responsible for about 40% of road accidents. Several misfortunes can be avoided if the driver is alerted in time.
Most of the time the drivers would lose their alertness and meet with unfortunate accidents. This loss of the state of alertness is due
to fatigue and drowsiness of the driver. This situation becomes very dangerous when the driver is alone. The ultimate reason for the
loss of the state of alertness is accidental microsleeps. Drowsiness or fatigue is one of the main reasons of low road safety and some
severe injuries, economy loss, and even deaths. Collectively, these situations increase the risk of road accidents. Using computer for
automatic fatigue detection, several misfortunes can be avoided. The drowsiness detection systems continuously analyses the
driver’s condition and warns before any unfortunate situation arises. Due to the accidents being caused due to the fatigue state of the
drivers, several methods have been developed to detect the driver’s drowsiness state and warn accordingly. Each method has its
advantages as well as disadvantages. There have been some great works in this field, but we can have some space for future
improvements. Late insights, assess that yearly 1,200 deaths and 76,000 injuries can be credited to weariness related accidents. It
can be seen there are around 2,400 road accidents consistently which is one death per every four hours. It has been figured around
20% of car crashes with driver fatalities are due to driver’s drowsiness. It was uncovered that driving execution quickly drop with
expanded tiredness which result in making more than 20% of all vehicle accidents. Less attention heads the driver to being
distracted and the likelihood of street accident goes high. Drowsiness related accidents have all the earmarks of being more serious,
because of the higher speeds involved distraction and the driver being not able to take any avoiding activity, or even brake, before
the accident. The improvement of innovations for recognizing or preventing tiredness of the driver is a significant test in the field of
accident preventing systems. Because of the danger that that drowsiness presents on the road, strategies need to be created for
checking its influences. Loss of the awareness because of the tiredness causes a few changes in the human's body and activities.
These side effects and parameters empower us to effectively measure the drowsiness level. Because of the hazard that drowsiness
presents on the road, methods need to be developed for counteracting its affects. Driver inattention might be the result of a lack of
alertness when driving due to driver drowsiness and distraction. Driver distraction occurs when an object or event draws a person’s
attention away from the driving task. Unlike driver distraction, driver drowsiness involves no triggering event but, instead, is
characterized by a progressive withdrawal of attention from the road and traffic. Additionally, we believe that drowsiness can
negatively impact people in working and classroom environments as well. Although sleep deprivation and college go hand in hand,
drowsiness in the workplace especially while working with heavy machinery may result in serious injuries similar to those that
occur while driving drowsily. 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
©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 129
II. LITERATURE REVIEW

A survey done by National Highway Traffic Safety Administration estimated that there were 56,000 sleep related road crashes in the
U.S.A in 1996. Another survey done in 2007 says that 18% of accidents involved fatigue as the main factor. In Britain up to 20% of
serious road accidents were caused due to fatigue. Similarly, survey done by the Road and Traffic Authority states that in the year
2007, fatigue contributed to 20% of accidents caused on road. Accidents due to drowsy was prevented and controlled when the
vehicle is out of control. The term used here for the identification that the driver is drowsy is by using eye blink of the driver.
These types of accidents occurred due to drowsy and driver could not be able to control the vehicle, when the driver wakes. The
drowsiness was identified by the eye blink closure rate. If the driver is in drowsy state, then the system will give buzzer signal.
One of the suggested methods is to monitor the movement of the vehicle to detect drowsiness of the driver. However this method
has limitations as the results are influenced by the type of vehicle and the condition of road. Another method is to process the
electrocardiogram (ECG) signals of driver. This approach also has limitations as ECG probes shall always be connected to the
driver‘s body. That would disturb the driver. Few researches tried to assess the fatigue factor by monitoring the eye blink rate of the
driver. Successful detection of eye blink rate has been the interest of many researchers proposed methods based on combination of
projection and the geometry feature of iris and pupil.
Some works also are based on "Support Vector Machine" (SVM) classifier. The SVM classifier is used to detect state of the eye.
SVM classifier and Gabor filter has been used to extract eye characteristic. In the above methods, the authors used some conditions
which make some difficulties in the eye state recognition. The system detects the fatigue symptoms of the driver which consists of
an eye blink sensor for driver blink attainment and an adaptive speed controller designed using stepper motor for providing actual
positioning of the throttle valve to adjust the speed of vehicle. Advanced technology offers some hope to avoid these up to some
extent. This paper involves measure and control of accidents by using both alcohol sensor and IR sensor.
It uses remotely located charge-coupled-device cameras with active infrared illuminators to acquire video images of the driver.
Various visual signs that typically characterize the level of alertness of a person were extracted in real-time and systematically
combined to infer the fatigue level of the driver. The visual cues employed characterize eyelid, gaze, head movements and facial
expressions. A probabilistic model was developed to model human fatigue and to estimate fatigue based on the visual cues. The
simultaneous use of visual cues and their systematic combination earns an accurate fatigue characterization. This system was
validated under real-life fatigue conditions with the human subjects of different ethnic backgrounds with or without glasses; and
beneath different illumination conditions. It was found to be reasonably reliable, and accurate in fatigue characterization.
The computer vision based method determine if a driver is holding a cell phone close of his/her ears using the Supervised Descent
Method (SDM) which it tracks some facial landmarks to extract a crop of regions of interest (ROI) (the driver‘s ear region).
Features are extracted from the ROIs and the phone usage is detected using previously trained classifiers. The system can be
processed in near real time. The approach of Yang send beeps of high frequency through the car sound equipment, network
Bluetooth, and use software running on the phone for capturing and processing sound signals. The beeps are used to calculate the
position where the cell phone is, and then we know when the driver (or another passenger in the car) is talking on it. The proposal
achieved a classification accuracy of more than 90%. This approach works with hands-free usage, but it depends on the operating
system and mobile phone brand, and the software has to be continually enabled by the driver.
Another proposal identifies the behaviours of a distracted driver associated with text messaging. The approach uses a cell phone
programmed to record any typing done (pressing and releasing any key). An analysis can be performed to verify distractions
through these records. Experiments were done with six participants used the cell phone as passenger and driver what distinct
patterns of typing frequency were shown in each situation. The driver cannot reply the text messaging with an average frequency of
2 press keys by the second. This constraint detects the driver in 99% of cases, but it works with offline data processing.
The system could be applied to others equipment, i.e, the GPS.
Another system presents for detecting and parsing the movement of the driver‘s head. It uses the optical flow of the driver‘s image.
The detected movements are compared with a predefined set of relevant movements of the automotive environment. The system
observes the time spent which the driver is looking in the same direction. If the time spent is longer than a predefined interval of
time, the distraction is observed. The system‘s accuracy in a real environment is 86%. The approach finds the driver distraction level
for observing objects outside of the vehicle. A fusion of two computer vision systems is used.
The first one detects the driver‘s field of view (inside of the vehicle) and the other detects the movement (outside) by using the
salience map (the outside movements what must attract the driver‘s attention).
Over the last decade, there have been various studies done related to drowsiness detection and drunk driving. Features using a
driver‘s Visual characteristics, Physiological and Driving- behaviour based studies have been conducted each having their own
advantages for drowsiness detection and by using sensors for drunk driving detection. A survey that provides a comprehensive
insight into the well-established techniques for driver inattention monitoring and introduces the use of most recent and futuristic
solutions exploiting mobile technologies such as smartphones and wearable devices. The studies were categorized into two groups:
driver drowsiness and distraction. A comprehensive compilation, used features, classification methods, accuracy rates, system
parameters, and environmental details, was represented.
A similar approach was also taken for the methods used for the detection of driver distraction. A visual analysis of Eye State and
Head Pose (HP) for continuous monitoring of alertness of a vehicle driver. The proposed scheme used visual features such as Eye
Index (EI), Pupil Activity (PA), and HP to extract critical information on non-alertness of the driver . A system is designed and
implemented an automatic system, using computer vision, which runs on a computationally limited embedded smart camera
platform to detect yawning. Implementation of the Viola–Jones algorithm for face and mouth detections and, use of a back-
projection theory for measuring both the rate and the amount of the changes in the mouth, to detect yawning along with the
histogram of the Gray scale image.
A warning alarm was also sounded if driver fatigue was believed to reach a defined threshold. And presented an efficient driver‘s
drowsiness detection system, by using yawning detection. The consideration of eye detection and mouth detection was done,
detecting the driver‘s face using YCbCr method. After that, eyes and mouth positions by using HAAR features. Lastly yawning
detection performed by using mouth geometric features.
There has been conducted the research in a relatively simpler way. They claim the sleep onset is the most critical consequence of
fatigued driving, separate the issue of sleep onset from the global analysis of the physiological state of fatigue, and take eyes
opening and closing as cues of sleep onset. They have used vision-based system to monitor the eyes conditions in order to detect
fatigue in driving. Lee et al. have used two fixed cameras to capture the driver‘s sight line and the driving lane path for the purpose
of driving pattern and status recognition. They calculate the correlation coefficients among them to monitor the driving status and
patterns. These methods all need one or more cameras to be installed in the vehicle and just in front of the driver. It will cause
certain potential safety hazard to the driver
III. LITERATURE SUMMARY
Sl. Year of Methodology /

Paper title, Author Results obtained
No Publication Algorithms used
Haar cascade
algorithm is used
to see the driver’s
Driver Drowsiness The performance in real
face and eye
Detection Using Deep time is excellent. The
regions. Then, the
Learning, Ajinkya driver drowsiness system
1 2021 system is trained
Rajkar, Nilima works with an average
with the proposed
Kulkarni, and Aniket accuracy of 96%.
convolutional
Raut
neural network
for the detection
of drowsiness.
Convolutional
Neural Network
Distracted Driver
(CNN), High- The results indicate a
Detection with Deep
level deep positive outlook on re-
Convolutional Neural
learning package purposing the VGG16
Network, O. G. November 2019
2 known as Keras is model to classify
Basubeit, D. N. T. How,
extensively used distracted drivers with up
Y. C. Hou, K. S. M.
for rapid to 96% accuracy on
Sahari
experimentation unseen images
of pretrained
CNN models.
Their result proves that

Real-Time Driver State
temporal information
Monitoring Using a
provides a considerable
CNN Based Spatio-
improvement in
Temporal CNN Based
18 Jul 2019 classification accuracy.
3 Approach, Neslihan Spatio-Temporal
this approach also
Kose1, Okan Kopuklu2 , Approach
provides real-time
Alexander Unnervik3
performance while
and Gerhard Rigoll4
detecting distracted
drivers
They
demonstrated
three different Experimental results
techniques: clearly depict that
Distracted Driver
applying these
Detection using CNN 1. Simple CNN on
techniques has increased
and Data Augmentation image,
April 2018 the accuracy of their
4 Techniques, Vasanti 2. CNN applied on
model by decreasing
Sathe, Neha Prabhune, parts-based
overfitting. The approach
Anniruddha Humane augmented data,
is mostly useful when a
3. CNN applied on dataset with low
class-based variability .
augmented data
Based on real time data

Drowsiness
Drowsiness Detection of capturing and analysis,
Detection of a
a Driver using eye blinking and yawn
Driver using
Conventional Computer Feb 28-29, detection are considered
Conventional
5 Vision 2020 important parameters to
Computer Vision
Application,Hitendra detect drowsiness and
Application. They
Garg fatigue of the driver and
use Haar
ring the alarm
algorithm also.
accordingly.
Support Vector Up to 86.67% of the
Driver drowsiness
Machine (SVM) trials accurately detect
detection with eyelid
to detect the when the driver is sleepy.
related parameters by
drowsiness of the 16.67% of the trials
6 Support Vector 2018
drivers with which are supposed to be
Machine, Hu Shuyan *,
multiple eyelid alert are wrongly
Zheng Gangtie
movement detected as sleepy very
features. sleepy.
Real-time Driver Drowsiness
Detection for Android
Application Using Deep They have uses According to the
Neural Networks Techniques special facial experimental results, the
7 Rateb Jabbara*, Khalifa Al- 2018 features obtained size of the used model is
Khalifa, Mohamed through MLP. small while having an
Kharbechea , Wael Alhaj accuracy rate of 81%.
Yaseen , Mohsen Jafarib ,
Shan Jiangb
It shows that the proposed
They have uses approach outperforms the
Real-time Detection of
deep CNN baseline one which has only
Distracted Driving
architectures 256 neurons in the fully
based on Deep
including VGG- connected layers. and the
8 Learning, Weihua 2018
16 , AlexNet , results indicate that the
Sheng,Duy Tran,Ha
GoogleNet and Google Net is the best model
Do,he Bai
residual network out of the four for distraction
(ResNet) . detection in the driving
simulator testbed.
Detecting Driver It develop a system that

Drowsiness in Real benefits from the
Mobile Net CNN
Time Through Deep unmatchable accuracies
architecture with
Learning Based Object of convolutional neural
Single Shot
Detection, Nabit A. networks in computer
Multibook
9 Bajwa ,Ahmad 2019 vision tasks while at the
Detector (SSD) is
Muhammad same be computationally
used for this task
Anwaar ,Anabia Sohail resource efficient so that
of drowsiness
and Asifullah Khan it could be deployed to
detection.
cheaper embedded
devices.
Real-Time Driver State This result proves that
Monitoring Using a temporal information
CNN Based Spatio- provides a considerable
Temporal Approach, They have used a improvement in
Neslihan Kose1, Okan 18 Jul 2019 CNN Based classification accuracy.
10
Kopuklu2 , Alexander Spatio-Temporal This approach also
Unnervik3 and Gerhard Approach. provides real-time
Rigoll4 performance while
detecting distracted
drivers.
They have used a
Convolutional
Neural Network
Detection of Distracted method for With the accuracy of
Driver using detecting 96.31% the proposed
Convolutional Neural distracted driver. system outperforms
11 Network,Bhakti Baheti 2018 The pre-trained earlier approaches of
Suhas Gajre Sanjay ImageNet model distracted driver
Talbar is used for weight detection from literature
initialisation and on the dataset.
concept of
transfer learning
also applied.
It has averaged
the results of the 3
models namely
Distracted Driver VGG-16, VGG-
Detection and 19 and The best VGG-16 model
Classification ,Prof. InceptionV3 to had a log loss value of
Pramila M. get the final 0.8157. Similarly, VGG-
12 Chawan,Shreyas 2018 prediction values. 19 and InceptionV3 had
Satardekar,Dharmin They have use the log loss values of
Shah,Rohit SVM model to 0.9631 and 1.0972
Badugu,Abhishek Pawar detect the use of respectively.
mobile phone
while driving.
And also uses the
CNN model.
They have used
three deep
learning models,
namely, the
Distracted Driver
residual network
Classification Using It was able to achieve
(ResNet), the
Deep Learning, Munif very accurate results
13 2020 hierarchical
Alotaibi, Bandar particularly when there is
recurrent neural
Alotaibi sufficient training data.
network (HRNN),
and the Inception
architecture by
combining them
into one model.
It is built using a
The result, their system
commercial
provides the best
camera
Real-Time Driver- accuracy when the cab is
automobile device
Drowsiness Detection bright, and the driver
and a commercial
System Using Facial August 21, wears no glasses. If the
14 cell phone that
Features, WANGHUA 2019 driver wears glasses and
stores the result.
DENG1 AND the driving environment
And CNN to
RUOXUE WU is slightly dim, the
recognize the
accuracy of fatigue
eight layers of the
driving is reduced.
eye state
Real-Time Driver-
They have used
Drowsiness Detection
convolutional
System Using Facial convolutional neural
neural networks,
Features, WANGHUA network able to detect
Training and
DENG1 AND distracted drivers and
15 2019 testing is carried
RUOXUE WU Real- also the cause of
out using two
Time Distracted Drivers distraction with 95.82%
NVIDIA 1080 Ti
Detection Using Deep accuracy
GPUs with 11GB
Learning, Vlad Tamas,
RAM each.
Vistrian Maties
IV. METHODOLOGY
When people start driving, they tend to gaze at the left or right sides of the road after setting a destination. However, as time passes
by, the frequency of blinking decreases as a driver becomes fatigued. A reduced rate of blinking may result in drivers closing their
eyes unconsciously, and their heads may flop. These temporal behaviors can be observed continuously. Therefore, temporal
changes of facial features may be analyzed to identify changes in a driver’s condition.
In this section results are obtained using software and hardware platforms to achieving the objective of driver drowsiness detection
and alcoholic intoxication. Besides eye and head movements, another visual cue that can potentially capture one’s level of
drowsiness is his/her eyes and faces detection analysis. Making a real time application with computer vision is very effective and
efficient challenging task that needs processing powerful system. OpenCV is open source software, which is used for creating
computer vision. OpenCV is available in C, C++, and Python and Java programming languages extension. In case of driver is in
sleepy or finding fatigue, the message will be sent by using GSM and buzzer will be turned on till the GSM positive message from
car owner.. Fig. 1 shows the basic block diagram of the proposed system. Haar Feature based Cascade Classifier technique, it is a
machine learning based approach where a cascade function is trained from a lot of positive and negative images, and this positive
image is used for detecting face region and eye region the update of region of interest ROI. Open CV is packed with a trainer as
well as detector. The open CV is used for creating user defined object classifier. The object classifier that has been created is stored
in.xml file extension classifier can be used in the later stages of programming. Also in this paper we use canny operator edge
detection for recognize exact coordinate of eyes region Cellphone ring detector used here to detect cellphone usage during the
driving and warn the driver to restrict driver not to use mobile while driving.
V. CONCLUSION
Purpose of our project is to help solving real life problem in very cost-efficient way. It alerts the truck driver, whenever the driver
feels drowsy and closes his eyes for more than a second, As a result, it alerts the driver. As a result the accident ratio decreases.
Hence, our project if commercially developed will help in saving the precious life of driver. The system which can differentiate
normal eye blink and drowsiness which can prevent the driver from entering the state of sleepiness while driving. During the
monitoring, the system can decide if the eyes are opened or closed. The driver drowsiness system will reduce accidents causing and
provides safe life to the driver and vehicle safety. A system for driver safety and car security is presented only in the luxurious
costly cars. Using drowsiness detecting system, driver safety can be implemented in normal cars also.
REFERENCES
[1] Driver Drowsiness Detection Using Deep Learning, Ajinkya Ranker, Nilima Kulkarni, and Aniket Raut
[2] Distracted Driver Detection with Deep Convolutional Neural Network O. G. Basubeit, D. N. T. How, Y. C. Hou, K. S. M. Sahari international
Journal of Recent Technology and Engineering (IJRTE) ISSN: 2277-3878, Volume-8 Issue-4, November 2019
[3] Real-Time Driver State Monitoring Using a CNN Based Spatio-Temporal Approach, Neslihan Kose1 , Okan Kopuklu2 , Alexander Unnervik3 and
Gerhard Rigoll4, 18 Jul 2019
[4] Real-Time Driver State Monitoring Using a CNN Based Spatio-Temporal Approach, Neslihan Kose1 , Okan Kopuklu2 , Alexander Unnervik3 and
[5] Drowsiness Detection of a Driver using Conventional Computer Vision Application, Hitendra Garg, Feb 28-29, 2020
[6] Driver drowsiness detection with eyelid related parameters by Support Vector Machine Hu Shuyan *, Zheng Gangtie School of Astronautics, Beijing
University of Aeronautics and Astronautics, 37 Xueyuan Road, Haidian District, Beijing 100083, China
[7] Real-time Driver Drowsiness Detection for Android Application Using Deep Neural Networks Techniques, Rateb Jabbara* , Khalifa Al-Khalifa ,
Mohamed Kharbechea , Wael Alhaj Yaseen , Mohsen Jafarib , Shan Jiangb ,2018
[8] real-time Detection of Distracted Driving based on Deep Learning. Weihua Sheng,Duy Tran,Ha Do,he Bai,July 2018
[9] Detecting Driver Drowsiness in Real Time Through Deep Learning Based Object Detection Nabit A. Bajwa, Ahmad Muhammad Anwaar ,Anabia Sohail
and Asifullah Khan 2019
[10] Real-Time Driver State Monitoring Using a CNN Based Spatio-Temporal Approach* Neslihan Kose1 , Okan Kopuklu2 , Alexander Unnervik3 and
[11] Detection of Distracted Driver using Convolutional Neural Network Bhakti Baheti Suhas Gajre Sanjay Talbar,2018
[12] Distracted Driver Detection and Classification, Prof. Pramila M. Chawan,Shreyas Satardekar,Dharmin Shah,Rohit Badugu,Abhishek Pawar, 2018
[13] Distracted Driver Classification Using Deep Learning ? Munif Alotaibia,1 , Bandar Alotaibib,2020
[14] Received August 4, 2019, accepted August 18, 2019, date of publication August 21, 2019, date of current version September 5, 2019. Digital Object
Identifier 10.1109/ACCESS.2019.2936663 Real-Time Driver-Drowsiness Detection System Using Facial Features WANGHUA DENG1 AND RUOXUE
WU 1,2
[15] Vlad Tamas, Vistrian Maties. Real-Time Distracted Drivers Detection Using Deep Learning. American Journal of Artificial Intelligence. Vol. 3, No. 1,
2019, pp. 1-8. doi: 10.11648/j.ajai.20190301.11 Received: February 22, 2019; Accepted: April 8, 2019; Published: May 15, 2019

Driver Drowsiness Detection System

Uploaded by

Copyright:

Available Formats

Driver Drowsiness Detection System

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Driver Drowsiness Detection System

Uploaded by

Copyright:

Available Formats

10 V May 2022

Driver Drowsiness Detection System

II. LITERATURE REVIEW

III. LITERATURE SUMMARY

Sl. Year of Methodology /

Their result proves that

Based on real time data

Detecting Driver It develop a system that

You might also like