International Journal of Engineering &Technology, 7 (3.

12) (2018) 498-504

International Journal of Engineering & Technology

Website: www.sciencepubco.com/index.php/IJET

Research paper

Detection of Driver Drowsiness using Eye Blink Sensor

Kusuma Kumari B.M1, Sampada Sethi2, Ramakanth Kumar P3, Nishant Kumar4,
Atulit Shankar5
1
Assistant Professor, Department of Computer Science, University College of Science,
Tumkur University, Tumakuru, Karnataka
2, 4, 5
Student, 3Professor, Department of Information Science and Engineering,
R.V. College of Engineering, Bangalore, Karnataka
*Corresponding Author E-mail: [email protected], [email protected],
3
[email protected], [email protected]
5
[email protected]

Abstract

Accidents due to driver drowsiness can be prevented using eye blink sensors. The driver is supposed to wear the eye blink sensor frame
throughout the course of driving and blink has to be for a couple of seconds to detect drowsiness. Any random changes in steering
movement leads to reduction in wheel speed. The threshold of the vibration sensor can be varied and accordingly action can be taken.
The outcome is that the vibrator attached to eye blink sensor’s frame vibrates if the driver falls asleep and also the LCD displays the
warning messages. The wheel is slowed or stopped depending on the condition. This is accompanied by the owner being notified through
the GSM module, so the owner can retrieve the driver’s location, photograph and police station list near to driver’s location. This is how
the driver can be alerted during drowsiness and the owner can be notified simultaneously

Keywords: eye blink sensor, LCD, Microcontroller, Drowsy Driver Detection System

2. Literature Survey
1. Introduction
Automatic driver drowsiness can be detected using artificial
For any vehicle accidents driver’s faults are the most accountable intelligence and visual information.
aspect to cause dangerous problem to the society. Many drivers
cannot control the vehicles due to different reasons it may cause System is to detect, track and examine face and eyes of drivers for
severe accidents and sometime death. For vehicle accidents this different real vehicle image of drivers are taken to validate the
various factors involved such as drunk driving, over speeding, algorithms.
many distractions like texting while driving, talking with others, It is a real time system work in different light conditions [1]. The
playing with children etc. one of the important factor is sleeping numbers of accidents are increased due to several factor, one of
on the wheel. People know the dangerous of alcohol consumption the main factor is that driver fatigue. Driver’s sleepiness is also
and run the vehicles but they not understand the seriousness of implemented using video based approach. This system is
fatigue driving. In India, Ministry of Road Transport and Highway noninvasive and human related elements are used. Band power
released a report in 2015, every day around 1,374 accidents may and Empirical Mode Decomposition methods are used to
happen and almost 400 people deaths occur. Every hour because investigate and extract the signal, SVM (Support Vector Machine)
of vehicle accidents approximately 57 road accidents and 17 used to confirm the analysis and to categorize the state of
people dies. In that 54.1 percent of people are in the age group of vigilance of the driver [2].
15 to 34 years are killed in vehicle accident. The Government of The system designs to find the drivers drowsiness using the
India, Ministry of Road Transport and Highway Government of hypothesis of Bayesian networks. The interaction between driver
India prepare a strategy to diminish the amount of motorway and vehicle features are extracted to get reliable symptoms of
accidents and losses by 50 % by 2020. driver drowsiness. It presents more suitable and accurate strategies
Globally vehicle accidents have seemed one of the major to design drowsy driver detection system [3].
community health problems. In India almost 5 lakh road accidents Brain and visual activity is used in drowsiness detection system.
happened in the year 2015. A fatigue Driver those who falls asleep Electroencephalographic (EEG) channel used to monitor the brain
at the move fails to control the vehicle, not possible to take activity. Diagnostic techniques and fuzzy logic are used in EEG-
immediate action and results in a crash so it is necessary to based drowsiness detector. Using blinking detection and
monitor the drowsiness of the driver to prevent accidents. characterization for visual activity monitored. Electrooculographic
(EOG) channel are used to extract the Blinking features [4].
Image processing and pattern classification used to take the driver
facial pictures, tracking the features of driver face and
categorizing the driver’s sleepiness level. 17 different features
Copyright © 2018 Authors. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original work is properly cited.
International Journal of Engineering & Technology 499

points are determined after examining the facial muscle activities 3.6 L298 Motor Driver
using Active Appearance Model (AAM). K-Nearest-Neighbor
method applied to categorize sleepiness into 6 levels, driver’s The DC motor receives power from this motor driver, which in
smile also detected with this method [5]. turn is powered from the SST microcontroller board. The enable
Head posture estimation method is used for detection of drowsy pins of this driver should be high to send the power to the motor.
driver. In this method Viola and Jones algorithm for driver face
detection [6]. 3.7 GSM Module
This method is nonintrusive and sturdy for finding the driver
drowsiness in real time. Support Vector Machine (SVM) is using The GSM module used is GSM-SIM300. This module’s primary
for extracting the face from video frames and Circular Hough function is to send an audio warning message to the owner’s
Transform (CHT) is useful for mouth and eye state analysis [7]. registered number that is, in case of any accident the owner is
In this approach machine learning used to determine the human notified by it android application which plays an audio saying
behavior during driver drowsiness, for this 30 different facial ―accident occurred, kindly check your phone‖. A sim card is to be
actions including eye blink, yawning and head movements are fixed in this module to simultaneously send a text message on the
collected to detect the driver drowsiness [8]. owner’s phone.

3. Methodology 3.8 Accelerometer 3-Axis and ADC0809

The accelerometer used is ADXL335 3-axis which acts as steering

The concept of drowsy driver detection system focuses on the
and measures any increase in the speed. The reference speed taken
functioning of all sensor modules used in the project. This helps
in this project is 150kmph. As the vehicle turns randomly, the
explain the inputs received by modules and the outputs they
accelerometer generates analog values which are further given to
produce.
the ADC0809, the converter used in this project. These digital
values in turn act as inputs to the dc motor, the LCD and the
3.1 Eye Blink Sensor vibrator.
This sensor module consists of the eye blink sensor frame, the IR
sensor and a relay. The vibrator device is connected to the eye
3.9 SST Microcontroller
blink sensor frame which is to be worn by the driver. This vibrator
This is the main working unit which connects all the inputs and
vibrates whenever an accident occurs or the driver falls asleep.
outputs to its 4 ports namely, P0 P1 P2 P3 and each of them have
The frame consists of the IR transmitter which transmits the IR
8-pins which act as pins for input and output. The main ac dc
rays towards the driver’s eyes and an IR receiver which receives
power supply f 12V is given to this board. It has a huge RAM and
the reflected rays when the eyes are closed. The relay provides the
ROM which helps the program run faster as compared to other
extra current required by this module and hence is also connected
microcontrollers and also provides extra memory storage capacity
to the SST microcontroller board [9].
[11].
3.2 LCD
4. Implementation
LCD is a device which displays the messages in case the driver
falls asleep or accident occurs. The messages displayed are 4.1 Hardware Requirements
―SLEEPING‖ and ―ACCIDENT OCCURRED‖ as per the
situation. All the modules are connected to it so that the particular The various sensors form the hardware part of the project and
signals can be received and hence message could be displayed. It power supply is AC DC 12V adapter. Other hardware modules are
uses the power supply from the SST microcontroller board specified as follows:
displays a ―WELCOME‖ message. It also provides 5V power to 1) PIC16F877A.
other modules [10]. 2) CAN Transceivers
3) Alarm circuit.
3.3 LM317PSU Board 4) SST Micro Controller
5) Power Supply
This board is connected to the relay chip which in further 6) IR eye blink sensor
connected to the eye blink sensor system module. This board’s 7) LM 358 comparator
main purpose is to fulfill the power requirements of the eye blink 8) Accelerometer
sensor module. 9) LCD
10) Vibrator Output devices
3.4 DC Motor
4.2 Software Requirements
The motor acts as the wheel of the vehicle and it rotates when the
power is supplied to it through L298 chip. The speed of rotation is For the implementation of the modules, the coding language is
slowed down when the driver falls asleep as detected by the eye specified in this part and also the platforms used for coding and
blink sensor, in the other case the wheel is stopped when the the android application development:
accident occurs. 1) Keil C compiler
2) Embedded C
3.5 Vibration Sensor 3) Android studio

The purpose of this sensor is to sense any jerk given to the vehicle 4.3 General Constraints
which is the emulation of the accident occurrence in real time.
This receives power from the 5V port of the LCD. To developing the system the subsequent constraints were kept in
The output produces and sends signals to dc motor driver and mind:
stops the rotation of the wheel, that is, the motor.  The IR eye blink sensor’s transmitter and receiver should
be in a straight line.
500 International Journal of Engineering & Technology

 There shall be no element of doubt regarding the Drowsy driver detection system is divided into subsystems and
correctness of the distance measured. these sub systems gives some interconnected services.
 The system must have the android app for location Architecture design can be defined as proving a framework for
detection since phone’s processor is faster than that used by GSM sub-system control, in the first stage of the design process all
module. subsystems of the drowsy driver detection system are identified
 The system should be robust. and hence, the design process output is a output of this a
explanation of the software architecture. A basic structural
4.4 Development Methods framework for a system can be established through architectural
design process. The identification of the major components of the
The development methods decide a large section of how the final system and communications between these components is
system functions, and thus care is taken to ensure that the best considered under this.
practices, tools and equipments are used. The system will be The system architecture below portrays the blocks required for the
developed using Keil IDE. The programming language embedded implemented system. Figure 1 outlines the system architecture.
C will also be used. Certain sections of the system will be ported The sensors need to detect respective parameters and pass the
android application development for performance enhancement. signals to microcontroller. The outputs from SST are used as
The main modules would be microcontroller, IR sensor, LM358 inputs by various output devices to control the speed and display
comparator, accelerometer, LCD and android phone [12]. the message [13].

4.5 System Architecture

Fig. 1: System Architecture

Figure 2., depicts the structure chart of the complete system. The
structure chart contains all the components of the system.
International Journal of Engineering & Technology 501

Fig. 2: Structure Chart

 Input
4.6 Input Module The inputs to this module are signals from ports of
microcontroller.
This section is responsible for taking in crisp values of the vehicle,  Output
eye blink movements any vibration caused to vehicle due to The output of SST would be passed to other output devices so
accident. various functionalities could be carried out as demanded by
 Purpose project.
The purpose of this module is obtaining values from each sensor
and comparing them with reference values. 4.9 Aid for All 8051 Alternatives
 Functionality
The functionality of this module is to obtain the raw values and One of the most fastest growing Microcontroller Architecture are
pass signals to SST. 8051 alternatives, today nearly more than 400 device alternatives
 Input available with various silicon vendors. For large applications
The input to this module are the crisp values of vehicle speed, IR Philips 80C51MX architecture the New extended 8051 devices are
from eye blink sensor and values from shock due to accident. used because of the large storage capacity with several Mbytes.
 Output Keil also provides different Development tools for best support for
The output of this module is fed into SST80C51 microcontroller. these different variants are recorded below [14].

4.7 SST Microcontroller 4.10 Bit 80C51 5 V Low Power 64 Kb Flash

Microcontroller
This section contains three inputs and outputs to various output
devices. It receives the signals from the previous modules The P89V51RD2 is an 80C51 microcontroller with 64 kB Flash
 Purpose and 1kB of data RAM. These devices are designed to be drop-in
The purpose of this module is to pass the received values as input and software-compatible replacements for the popular
to output devices. P89C51RB2/RC2/RD2 devices. It supports both the In-System
 Functionality Programming (ISP) and In-Application Programming (IAP) boot
This module provides inputs to the output devices. codes with 12-clock (default) or 6-clock mode selection.it contains
 Input three 16 bit times/counters. In-System programming and In-
The inputs would be outputs received by input devices. Application Programming on chip flash user code memory is 64
 Output kb. Operating frequency in 12x mode is 0 MHz to 40 M Hz. Non-
The output of SST would be passed to other output devices so volatile data storage is the 128-B page erase for efficient use of
various functionalities could be carried out as demanded by code memory [16].
project.
4.11 PCA
4.8 Output Devices
The Programmable Counter Array (PCA0) gives enhanced timer
This section contains DC motor, LCD, vibrator, GSM module functionality and requiring less CPU involvement compare to the
which perform their respective functionalities. standard 8051 counter/timers. Dedicated 16-bit counter/timer and
 Purpose five 16-bit capture/compare modules are there in PCA. Each
The purpose of this module is to carry out respective capture/compare module has its own associated I/O line (CEXn)
functionalities such as slowing down the vehicle speed, alerting which is routed through the Crossbar to Port I/O when enabled
the driver, notifying the owner and displaying message on the [17]. They can be programmed in any of the following modes:
LCD. Rising or falling edge timer, Software timer, High speed output
 Functionality and PWM
This module takes the input from the SST microcontroller through
ports and carries out intended goals.
502 International Journal of Engineering & Technology

4.12 Accelerometer  The owner’s phone receives an audio warning message.

 The owner can retrieve the location of the driver by choosing
An accelerometer is an electromechanical device. Acceleration the ―location‖ option.
force will measure using this device, due to cause of gravity i.e. g  The owner can also get a list of driver’s nearby police
force it shows acceleration and it measures in g units. It is used in stations. With this, the photograph of the driver can be sent to
tilt-sensing applications as well as dynamic acceleration resulting owner’s email address as specified in the clamped phone.
from motion, shock, or vibration.  The wheel/motor is stopped as soon as an accident occurs,
The ADXL335 is a popular three-axis analog accelerometer IC, the vibrator in the eye blink sensor frame vibrates and displays a
which reads off the X, Y and Z acceleration as analog voltages. message on the LCD.
Complete 3-axis acceleration measurement gives the ADXL335,  When the driver falls asleep, the vibrator vibrates and the
within range ±3 g in the x, y and z axis. Analog voltages are LCD displays the message. Along with this, the vehicle speed is
proportional to the acceleration are the output signals of this automatically reduced.
module. Poly silicon surface-micro machined sensor and signal  When the accelerometer is tilted randomly, that acts as the
conditioning circuitry contains in ADXL335 [18]. steering, a message is displayed on the LCD the speed of the
vehicle is reduced.
4.13 Eye Blink Sensor  With all of the above mentioned, the android applications
send and receives details simultaneously.
To observe the solidness of the traffic, we will be placing a few
sets of IR transmitter & receiver sensors on the side of the roads.
On side IR transmitter device will be installed & right opposite to
6.2 Vibration Sensor Senses Vibration
the IR transmitter, an IR receiver will be positioned. This set of IR
transmitter & receiver will be placed at interim distance on the
road [19].
The IR transmitters are attached to supply, so that they will
transmit high signal all the time. The IR receivers are attached to
the comparator circuit, to get digital signals. A low power
operational amplifier LM324 IC has been used to build a
comparator circuit. Two set of LM324 IC has been used in this
project [20].

4.14 Vibration Sensor

The vibration sensor buffers a piezoelectric transducer. The

transducer generates the voltages, when the transducer moved
from neutral axis, distortion makes strain within the piezoelectric
element. Security purpose only this vibration sensor decorator is
designed and it produced alarm through movement or vibration,
and it sends a signal to control panel, a new type of multi
directional security vibration detector with multi directional
Fig. 3: Vibration Sensor senses vibration
detection also developed [21].
Figure 3 shows that when a jerk or vibration is given to the sensor,
5. Testing the vibration sensor is activated and the wheel is stopped. The
LCD displays a warning message as ―ACCIDENT OCCURRED‖
For unit testing the following contents are suggestive testing and the vibrator on the eye blink sensor frame vibrates
strategy in the system [22]: continuously.
 The design description characteristics and source code for
each module to be trialed are first reviewed.
 An evaluation committee is responsible for validation of the
Unit Test Plan.
 A trial "stub" is created to allocate input to or fetch output
from the trial module.
 The code is then compiled in the trial environment to check
for any anomalies.
 The trials are executed and data received out of the verified
software are compared with the expected, as recorded in the Unit
Test Plan.
 The code is retested and results are recorded, when an
updated version is available
 Final report form archived from that unit is said to have
cleared all trials.
 Every possible tweaks/ related to software requirements and
draft documents are provided.

6. Results

6.1 The Drowsy Driver Detection System

The results obtained by running the module are as follows:

International Journal of Engineering & Technology 503

6.3 Eye is shut 6.6 List of Nearby Police Station

Fig. 4: Eye is shut

Figure 4 shows that when the eyes are closed then the vibrator
vibrates and the speed of the wheel is decreased. The LCD
displays a ―SLEEPING‖ message.
Fig. 7: List of nearby Police Station
6.4 Location from Clamped Phone to Owner’s Phone
Figure 7 shows the list of nearby police stations to the accident
location of the driver which is sent to owner’s phone.

6.7 Clamped Phone’s Activity Screen

Fig. 8: Clamped phone’s activity screen

Figure 8 shows the activity screen of clamped phone where the

Fig. 5: Location from clamped phone to owner’s phone details of the owner are to be entered. The email address is entered
here so that the photograph could be sent to the owner’s email-id.
Figure 5 shows that location of the driver on the owner’s phone.
The owner first receives a text message containing latitude and 6.8 Photograph Sent to Owner’s Email-Id
longitude and then the location is displayed on google maps.

6.5 Location Entered to Retrieve Nearby Police Station

Fig. 9: Photograph sent to owner’s email-id

Figure 9 Depicts the photograph received in the email of the

owner when the owner chooses the photograph option.
Fig. 6: Location entered to retrieve nearby Police Station

Figure 6 shows that after receiving the latitude and longitude from 7. Conclusion
the clamped phone, those values are typed in the field shown in
the figure. Then the option is selected to get the list of police The proposed system helpful to avoid vehicle accidents because of
stations driver’s sleepiness using eye blink sensor, in this paper we study
504 International Journal of Engineering & Technology

and design the system for driver fatigue detection. If the driver Control and Automation, 2007. ICCA 2007. IEEE International
becomes drowsy the eye blink sensor’s frame vibrates attached to Conference on. IEEE, pg45-60, 2007.
the vehicle and also the LCD displays the warning messages and it [17] Ogawa, Kenji, and Mitsuo Shimotani, ―A drowsiness detection
system‖, Mitsubishi Electric Advance: pg13-16, 1997.
alerts the driver’s through alarm sound to avoid the road accidents.
[18] Yeo, Jung-hack. "Driver's drowsiness detection method of drowsy
The wheel is slowed or stopped depending on the condition. This driving warning system‖, U.S. Patent No. 6,243,015. Pg:55-70, 5
is accompanied by the owner being notified through the GSM June 2001.
module, so the owner can retrieve the driver’s location, [19] Flores, Marco Javier, José María Armingol, and Adl Escalera,
photograph and a list of nearby police stations. ―Real-time drowsiness detection system for an intelligent
vehicle‖, Intelligent Vehicles Symposium, 2008 IEEE. IEEE, pg-
50-60, 2008.
References [20] Wierwille, Walter W., et al, ―Research on vehicle-based driver
status/performance monitoring; development, validation, and
[1] MarcoJavier Flores, JoséMaría Armingol and Arturo de la Escalera, refinement of algorithms for detection of driver drowsiness‖, Final
―Driver Drowsiness Warning System Using Visual Information for report. No. HS-808 247, pg-30-33, 1994.
Both Diurnal and Nocturnal Illumination Conditions‖, Springer, [21] Yang, Ji Hyun, et al, ―Detection of driver fatigue caused by sleep
EURASIP Journal on Advances in Signal Processing, 2010. deprivation‖, Systems, Man and Cybernetics, Part A: Systems and
[2] Belhassen AkroutWalid Mahdi, ―A Blinking Measurement Method Humans, IEEE Transactions on 39.4: pg694-705, 2009.
for Driver Drowsiness Detection‖, Springer, Proceedings of the 8th [22] Bergasa, Luis M., et al, ―Real-time system for monitoring driver
International Conference on Computer Recognition Systems vigilance‖,Intelligent Transportation Systems, IEEE Transactions
CORES, pp 651-660, 2013. on 7.1: pg.63-77, 2006.
[3] Ji Hyun Yang, Zhi-Hong Mao, Member, IEEE, Louis Tijerina, Tom
Pilutti, Joseph F. Coughlin, and Eric Feron, ―Detection of Driver
Fatigue Caused by Sleep Deprivation‖, IEEE TRANSACTIONS
ON SYSTEMS, MAN, AND CYBERNETICS—PART A:
SYSTEMS AND HUMANS, VOL. 39, NO. 4, JULY 2009.
[4] Antoine Picot, Sylvie Charbonnier, Alice Caplier, ―On-Line
Detection of Drowsiness Using Brain and Visual Information‖,
Published in: IEEE Transactions on Systems, Man, and Cybernetics
- Part A: Systems and Humans ( Volume: 42, Issue: 3)
Page(s): 764 – 775, May 2012.
[5] Satori Hachisuka, Kenji Ishida, Takeshi Enya, Masayoshi Kamijo,
―Facial Expression Measurement for Detecting Driver
Drowsiness‖, springer, International Conference on Engineering
Psychology and Cognitive Ergonomics. EPCE 2011: Engineering
Psychology and Cognitive Ergonomics pp 135-144, 2011.
[6] Ines Teyeb, Olfa Jemai, Mourad Zaied, Chokri Ben Amar, ―A
Drowsy Driver Detection System Based on a New Method of Head
Posture Estimation‖, springer. International Conference on
Intelligent Data Engineering and Automated Learning. IDEAL
2014: Intelligent Data Engineering and Automated Learning –
IDEAL 2014 pp 362-369, 2014.
[7] Nawal Alioua, Aouatif Amine, Mohammed Rziza,
Driss Aboutajdine, ―Driver’s Fatigue and Drowsiness Detection to
Reduce Traffic Accidents on Road‖, springer. International
Conference on Computer Analysis of Images and Patterns, CAIP:
Computer Analysis of Images and Patterns pp 397-404, 2011.
[8] Esra Vural, Mujdat Cetin, Aytul Ercil, Gwen Littlewort,
Marian Bartlett, Javier Movellan, ―Drowsy Driver Detection
Through Facial Movement Analysis‖, springer, International
Workshop on Human-Computer Interaction HCI 2007: Human–
Computer Interaction pp 6-18, 2007.
[9] Lin, Chin-Teng, et al. ―Drowsiness estimation for safety driving
using independent component analysis‖, Circuits and Systems I:
Regular Papers, IEEE Transactions on 52.12: 2726-2738 pg-20-30,
2005.
[10] Clarke Sr, James Russell, and Phyllis Maurer Clarke, ―Sleep
detection and driver alert apparatus‖, U.S. Patent No. 5, 689, 241,
pg25-70 18 Nov. 1997.
[11] Hayami, Takchito, et al. ―Detecting drowsiness while driving by
measuring eye movement-a pilot study‖, Intelligent Transportation
Systems, Proceedings. The IEEE 5th International Conference on.
IEEE, 2002 pg30-35, 2002.
[12] Hu, Shuyan, and Gangtie Zheng, ―Driver drowsiness detection with
eyelid related parameters by Support Vector Machine‖, Expert
Systems with Applications 36.4, pg651-658, 2009.
[13] Ito, Takehiro, et al, ―Driver blink measurement by the motion
picture processing and its application to drowsiness detection‖,
Intelligent Transportation Systems, Proceedings. The IEEE 5th
International Conference on. IEEE, 2002 pg30-35, 2002.
[14] Smith, Paul, Mubarak Shah, and Niels Da Vitoria Lobo,
―Determining driver visual attention with one camera‖, Intelligent
Transportation Systems, IEEE Transactions on 4.4: pg205-218,
2003.
[15] Kithil, Philip W., Roger D. Jones, and M. Jone, ―Development of
driver alertness detection system using overhead capacitive sensor
array‖, SAE Technical Paper Series 982292: pg35-56, 1998.
[16] Hong, Tianyi, Huabiao Qin, and Qianshu Sun, ―An improved real
time eye state identification system in driver drowsiness detection‖,