SLEEP REFLEXER
A PROJECT REPORT
Submitted by
* Aadhithya PB
* Avinas RM
* Azarudeen J
* Chandru A
* Emmanuel Louis L
in partial fulfillment for the award of the
degree of
BACHELOR OF ENGINEERING
in
ELECTRICAL AND ELECTRONICS ENGINEERING
SARANATHAN COLLEGE ENGINEERING, PANJAPUR
ANNA UNIVERSITY: CHENNAI 600 025
MAY 2022
ANNA UNIVERSITY : CHENNAI 600 025
BONAFIDE CERTIFICATE
Certified that this project report “SLEEP-REFLEXER” is the bonafide work
of “[Link] ,[Link], AZARUDEEN.J
CHANDRU.A, EMMANUEL LOUIS.L”
who carried out the project work under my supervision.
SIGNATURE SIGNATURE
HEAD OF THE DEPARTMENT
SUPERVISOR
<<Academic Designation>>
<<Department>>
<<Department>>
<<Full address of the Dept & College >> <<Full address of the Dept & College >>
[Link] TITLE [Link]
1 ABSTARCT 2
2 ACKNOWLEDGEMENT 4
3 INTRODUCTION 5
4 COMPONENTS USED 6
5 Block Diagram 16
6 CIRCUIT DIAGRAM 18
7 WORKING 19
8 PROGRAMMING CODE 24
9 PROTOTYPE IMAGES 26
10 LITERATURE REVIEW 28
11 ADVANTAGES AND 30
APPLICATIONS
12 FUTURE SCOPE 30
13 CONCLUSION 31
14 REFERNCE 32
ABSTRACT
In recent days, accidents occurring due to carelessness of
drivers has risen rapidly, major part of which is due to drivers falling
asleep while driving. In order to prevent this, we have designed a
system named “Sleep Driving Restrictor” module.
This paper presents an automatic drowsy driver monitoring and
accident prevention system that is based on monitoring the changes in
the eye blink duration. Our proposed method detects visual changes in
eye locations using the proposed horizontal symmetry feature of the
eyes
If the driver falls asleep, the vibrator attached to the eye blink
sensor's frame vibrates, and the wheel is slowed or stopped depending
on the situation. This is how the motorist can be notified if he or she
is drowsy.
Accidents due to drowsiness can be controlled and prevented
with the help of eye blink sensor using IR rays. It consists of IR
transmitter and an IR receiver. The transmitter transmits IR rays into
the eye. If the eye is shut, then the output is high.
Eye blink sensors can help reduce accidents caused by drowsy
drivers. The driver must wear the eye blink sensor frame for the
duration of the drive, blinking for a couple of seconds to identify
tiredness. Any erratic steering movement results in a drop in wheel
speed. The vibration sensor's threshold can be adjusted, and action
done as a result.
ACKNOWLEDGEMENT
Many people have aided me in finishing this project effectively.
I'd like to express my gratitude to everyone involved in this initiative.
First and foremost, I want to express my gratitude to
Dr. M. Marimuthu for allowing me to complete this project
successfully and I'll express my gratitude to Mrs. Suganyadevi for
instructing us throughout the project. Their advice and
recommendations aided in the accomplishment of this project.
Also I'd like to express my gratitude to my parents and friends
for providing me with invaluable advice and suggestions throughout
the project's development.
INTRODUCTION
Driver error is the most responsible part of any automobile
accident that causes a dangerous problem in society. Many drivers are
unable to manage their automobiles for various reasons, which can
result in serious accidents and even death. Major causes contribute to
car accidents, including drunk driving, excessive speeding, numerous
distractions. One of the most crucial factors is napping behind the
wheel. People are aware of the dangers of drinking and driving, but
most are unaware of the significance of tiredness driving.
According to a data provided by India's Ministry of Road
Transport and Highways in 2015, over 1,375 accidents occur every
day, resulting in nearly 397 deaths.
Vehicle accidents have been identified as one of the most
serious public issue in the world. In the year 2015, about 5 lakh
people died in traffic accidents in India.
A fatigued driver who loses control of the car, is unable to react
quickly, and crashes, hence it is vital to check the driver's drowsiness
to avoid accidents.
COMPONENTS USED
The drowsy driver detection system's concept focuses on how
all of the project's sensor modules work together. This clarifies the
inputs that modules receive and the outputs that they make.
● Eye blink sensor
● Arduino-nano
● Relay
● Buzzer
● DC motor
● Adaptor
Eye Blink Sensor
FUNCTION:
The eye blink sensor emits infrared waves that are reflected by the
receiver and detected. The sensor's output increases as soon as the eye
blinks. The Arduino board receives this output. To produce the noise,
the Arduino sends a signal to the buzzer.
DESCRIPTION:
There are three pins on it: a Vcc (supply) pin, a ground pin,
and a signal pin. This eye blink sensor produces a digital output.
The eye blink sensor frame, an IR sensor, and a relay make
up this sensor module. The eye blink sensor frame that the driver will
wear is coupled to the vibrator gadget. When an accident occurs or the
driver falls asleep, this vibrator vibrates.
The frame is made up of an infrared transmitter that sends IR
rays towards the driver's eyes and an infrared receiver that receives
reflected rays while the driver's eyes are closed. The relay, which is
also attached to the SST microcontroller board, supplies the extra
current required by this module.
Arduino-nano
DESCRIPTION
The Arduino Nano is a small, complete, flexible, and
breadboard-friendly Microcontroller board based on the
ATmega328p. It was developed by [Link] in Italy in 2008 and
includes 30 male I/O headers configured in the DIP30 style.
There are 14 digital pins, 8 analogue pins, 2 reset pins, and 6 power
pins on the Arduino Nano.
The Arduino IDE, which can be downloaded from the Arduino
website, is used to programme it.
The Arduino Nano is basically a smaller version of the Arduino UNO,
and the two boards have nearly identical functionality.
It has a 5V operational voltage, however the input voltage can be
anywhere between 7 and 12V.
The maximum current rating of the Arduino Nano is 40mA, so the
load connected to its pins should not drain more than that.
Each of these Digital and Analog Pins has a specific function.
[Link] is an open source MC board based on the
ATmega328P microcontroller. Digital pins and analogue pins are
available on the board. All of this will aid the microcontroller by
allowing it to be connected to a computer and run continuously. This
board's power can be supplied via an AC to DC converter, USB cable,
or plug.
FUNCTION
The Arduino Nano has a crystal oscillator with a frequency of
16 MHz. It's utilised to make a clock with a precise frequency by
employing a continuous voltage source. The Arduino Nano has one
drawback: it lacks a DC power connection, which means you can't use
it to power an external power source like a battery.
Name Arduino® Nano
Board
SKU A000005
Microcontroller ATmega328
USB connector Mini-B USB
Built-in LED Pin 13
Digital I/O Pins 14
Pins
Analog input pins 8
PWM pins 6
UART RX/TX
I2C A4 (SDA), A5 (SCL)
Communication
D11 (COPI), D12 (CIPO), D13 (SCK). Use
SPI
any GPIO for Chip Select (CS).
I/O Voltage 5V
Input voltage
7-12V
Power (nominal)
DC Current per
20 mA
I/O Pin
Clock speed Processor ATmega328 16 MHz
Memory ATmega328P 2KB SRAM, 32KB flash 1KB EEPROM
Relay
Description
This is a single channel relay shield, you can now turn on and off
lamps, fans, solenoids, and other small appliances that run on up to
250VAC or DC power.
1. The 8550 transistor drive, driveability.
2. A fixed bolt holes for easy installation.
3. It has a relay status indicator led Power LED(Green), 1
relay status indicator LED(Red)
4. Relay control interface by single-chip IO.
5. Low-level suction close, high-level release.
6. Easy to use, simple 3 line structure.
Function
A relay is an electromagnetic switch operated by a relatively small
current that can control much larger current. Here’s a simple
animation illustrating how the relay uses one circuit to switch on
another circuit.
Initially the first circuit is switched off and no current flows through it
until something (either a sensor or switch closing) turns it on. The
second circuit is also switched off.
When a small current flows through the first circuit, it activates the
electromagnet, which generates a magnetic field all around it.
The energized electromagnet attracts a contact in the second circuit
toward it, closing the switch and allowing a much bigger current to
flow through the second circuit.
When the current stops flowing, the contact goes back up to its
original position, switching the second circuit off again.
Arduino code
Buzzer
DESCRIPTION
Rated Voltage : 6V
DC Operating Voltage : 4 to 8V
DC Rated Current : ≤30Ma
Sound Output at 10 cm : ≥85dB
Resonant Frequency : 2300 ±300Hz
Tone : Continuous Operating Temperature : -25°C to +80°C Storage
Temperature : -30°C to +85°C Weight : 2g
The "Piezoelectric Sound Modules" provided here are based on the
principle of natural piezoelectric ceramic oscillation for conversion.
Peizo offers these buzzers in a variety of lightweight, portable sizes,
ranging from a modest 12 mm diameter to huge electrical outlets.
FUNCTION
The magnetic field attracts the vibrating disc in a magnetic buzzer to
the pole. When a signal oscillates through the coil, it creates a
fluctuating magnetic field that vibrates the disc at the same frequency
as the driving signal.
When engaged, it emits a continuous beep to wake the driver up when
he falls asleep for the first time. The Eye-Blink Sensor will be
connected to the buzzer.
BO Motor
DESCRIPTION
Rated Voltage: 3~6V
Continuous No-Load Current: 150mA +/- 10%
Min. Operating Speed (3V): 90+/- 10% RPM
Min. Operating Speed (6V): 200+/- 10% RPM
Torque: 0.15Nm ~0.60Nm
Stall Torque (6V): [Link]
Gear Ratio: 1:48
Body Dimensions: 70 x 22 x 18mm
Wires Length: 200mm & 28 AWG
Weight: 30.6g
Product Weight: 30.6g / 1.1oz
FUNCTION
An electric motor that transforms mechanical energy to
electrical energy is known as a DC motor.
The DC engine was the first extensively utilized into its field, a DC
car's speed can be modified to a greater range. Cars, toys, and
electrical appliances all employ small DC engines.
The universal motor has a lightweight brush engine and can operate
accurate action. It also has integrated power tools and devices.
Electric vehicles, elevator and hoist propulsion, and rolling steel
drivers all employ large DC engines. Instead of an automobile, we
useda DC motor in our project for practical reasons.
The motor revolves like a car axle as power is delivered to it.
Arduino UNO is an open source MC board created by [Link]
based on the ATmega328P MC. Pre-processing the Arduino system to
transform the design into a C++ system is part of the merging process.
It will then be forwarded to a moderator, who will convert the
intelligible human code into computer commands.
ADAPTOR
An AC adapter, also known as an AC/DC adapter, AC converter, or
charger, is an external power supply that is used with devices that are
powered by batteries or have no other source of power. By removing
the requirement for a standard sized power supply.
Each AC adapter has a specified power rating that it can handle and
output to an electronic device, measured in volts or watts. Because of
the power rating and plug type
Block Diagram
The components used in the proposed operation are Eye blink length
and frequency, Power supply, Buzzer, LED ARDUINO (UNO),
Relay Module, DC, and the system block diagram includes: Eye blink
(IR): related to sleep detection and alert the driver with the
components used in the proposed operation are Eye blink length and
frequency, Power supply, Buzzer, LED ARDUINO (UNO), Relay
Module, DC. The Arduino Uno is the core component, which is an
ATmega328-based microcontroller (MC) that handles all of the
embedded system circuit's functions.
The blinking module operates by illuminating the eye area with
infrared light, then utilizing an image transistor and a separation
circuit to modify the scattered light.
The eye blink sensor is used to detect eye blinks and can also be
used to identify behaviours such as the driver's drowsiness while
driving. It operates using infrared LED technology.
It includes an infrared transmitter and receiver LED for
detecting eye blinks. The following diagram depicts the operation of a
simple IR sensor:
Infrared sensors are made up of two parts, as indicated in the
image above: an infrared transmitter that serves as the source and an
infrared receiver that serves as the receiver.
IR LEDs are used as infrared sources, and photodiodes are used
as infrared detectors. The infrared source's radiation is reflected by an
item and returns to the infrared detector.
CIRCUIT DIAGRAM
WORKING
The RF Transmitter and Receiver sections make up the Driver
Drowsiness Detector. For data processing, the transmitter section
employs an RF Transmitter and an Eye Blink Sensor, while the
reception section uses an Arduino Uno with an RF receiver.
We previously utilised the same 433 MHz RF modules with Arduino
to create projects such as an Arduino RC Boat and a hand gesture
driven robot, among others.
The transmitter and receiver sections' circuit diagrams are shown
below.
TRASMITTER SIDE
The 9V DC battery is first stepped down to 5V DC using a 7805
voltage regulator, and then the 5V DC supply is delivered to the Eye
Blink Sensor and RF Transmitter, as indicated in the diagram.
The RF transmitter receives the eye blink sensor's output pin and
transmits it wirelessly to the receiving end.
RECIEVER SIDE
The RF receiver is connected to a 5V DC power source from Arduino
on the receiver side, as indicated in the diagram.
An external 12V DC power supply is used to power the Arduino. The
RF receiver's output is connected to the Arduino Analog pin. As seen,
the buzzer is linked to the Arduino's Digital pin.
The voltage drop across the series resistor/potentiometer will be large
when the incoming radiation is higher on the photodiode. Both the
reference analogue voltage and the actual output voltages are
compared in the Comparator IC, which is nothing more than an
Operational Amplifier, or Op-amp. The output of the comparator is
high if the voltage across the resistor series to the photodiode is
greater than the reference voltage, otherwise it is low. When the
sensor detects any action, such as eye blinking, the comparator's
output is coupled to an LED, which lights. Depending on the ambient
circumstances, the threshold voltage can be modified by adjusting the
potentiometer.
Eye Blink Sensor Technical Specifications
5V DC (Working Voltage)
TTL(5V/0V) output
Onboard Connections are made with a three-pin header.
The use of infrared technology
RF Transceiver Module at 433 MHz
Radio Frequency is abbreviated as RF. The frequency range related to
this is 30 kHz to 300 GHz. We're utilising a 433 MHz RF Transceiver
Module in this example. A 433 MHz RF Transmitter and RF Receiver
are included in this RF module. The transmitter/receiver (Tx/Rx) pair
uses a 433 MHz frequency. An RF transmitter receives serial data and
wirelessly delivers it through RF via its antenna. An RF receiver
operating at the same frequency as the transmitter receives the
transmitted data.
Radio Frequency Transmitter
The RF transmitter module runs at 433MHz and …
Ground: This is the ground of the transmitter. Attach to the ground
plane.
Input pin for serial data
VCC: Supply voltage; 5V ANT: RF Receiver Antenna Output Pin
The data is received by the RF receiver module and sent to the data
OUTPUT pin. The Microcontroller can decode the output data and
take action with it.
The receiver's ground is the receiver's ground. Attach to the ground
plane.
Data: Output pin for serial data
ANT: Antenna output pin; VCC: Supply voltage; 5V
Drowsiness Detector Testing CODE
void setup()
[Link](9600);
pinMode(2,OUTPUT);
}
The RF receiver's serial data output is read from analogue pin A0 and
stored in a variable as illustrated. A serial monitor can be used to
examine the reference values.
int x=analogRead(A0);
[Link](x);
To identify drowsiness, an if-else statement is written after calculating
the reference value. The millis() function is used to determine the
duration of the eye blink, as demonstrated in the code below. It beeps
a siren to inform the driver if it is more than specified value. In our
situation, the time length is 3000 milliseconds (1 second)
if(x<400 && flag==0) {
flag=1;
t1=millis();
else if(x>400 && flag==1) {
flag=0;
t2=millis();
[Link](t2-t1);
if((t2-t1)>1000) {
digitalWrite(2,HIGH);
[Link]("Alert!!!!!!!!!!!!!!!!!");
delay(2000);
digitalWrite(2,LOW);
}
else;
Now that the programme has been successfully uploaded to Arduino,
it's time to put it to the test. First, turn on the power on both the
transmitter and the receiver. Ensure that electricity is supplied from
both ends. Put on the sunglasses with the eye blink sensor and blink
your eyes naturally; an LED should light up on the sensor. If
unsuccessful, try adjusting the sensitivity of the potentiometer until it
correctly detects the eye blink. After successful identification, test for
tiredness by delaying blinking; the buzzer should beep to inform you.
Drowsiness Detector Testing for Drivers
That's all there is to it for creating a Sleep Sensing and Alerting
System for Drivers with an Arduino Nano and an Eye Blink Sensor.
Below is the complete code.
CODE
int flag=0;
int t1=0;
int t2=0;
void setup()
[Link](9600);
pinMode(2,OUTPUT);
}
void loop()
int x=analogRead(A0);
//[Link](x);
if(x<400 && flag==0)
flag=1;
t1=millis();
else if(x>400 && flag==1)
flag=0;
t2=millis();
[Link](t2-t1);
if((t2-t1)>1000)
digitalWrite(2,HIGH);
[Link]("Alert2!!!!!!!!!!!!!!!!!");
delay(3000);
digitalWrite(2,LOW);
else;
PROTOTYPE IMAGES
LITERATURE REVIEW
The author Mohamed sahal said that the eye-blink detector works by
employing infrared to illuminate the attention and/or lid space, then
using a phototransistor and differentiator circuit to monitor changes in
the reflected light.
The exact functioning is very dependent on how the emitter and
detector are positioned and aimed in relation to the eye.
Connect a 5 volt regulated DC power supply. The black wire is the
ground, the middle wire is the output, and the red wire is the positive
supply.
On the PCB, these wires are also labelled. To test the detector,
simply connect two wires (+5V and GND) to the detector. You may
leave the output wire alone because it is already connected. When the
eyelids are closed, the output is 0V. Place the Eye Blink Sensor Glass
on your face at a distance of 15mm, and you'll be able to detect
blinks.
The author suhas katkar and Mahesh manik kumbar said that The
Eye Blink sensor is based on infrared technology. As the eye blinks,
the variation throughout the eye will change. If the eye is closed, the
output is high, and if it is open, the output is low.
This is used to determine whether the eye is shutting or opening. This
signal is sent to a logic circuit, which signals the alarm. This can be
employed in a project that requires controlling an accident caused by
unconsciousness by blinking the eyes.
The IR receiver does not conduct when the IR rays between the IR
transmitter and receiver are interrupted. As a result, the comparator's
non-inverting input terminal voltage is greater than the inverting input
terminal voltage.
The comparator output is now in the +5V range. This voltage is
applied to the microcontroller or computer, causing the led to light.
The author Aysha Banu and Amaravathi said that the infrared sensor
CNY 70 was employed in this experiment. It is made up of an
infrared transmitter, which is a type of LED that emits infrared rays
and is known as an IR transmitter. Similarly, an IR Receiver receives
the IR rays sent out by an IR transmitter.
One crucial factor to remember is that the IR transmitter and receiver
should be in a straight line. When the sent signal is strong, it is sent to
the IR transmitter, which conducts the IR rays and sends them to the
receiver.
A comparator is attached to the IR receiver. The operational amplifier
LM 2903 is used to build the comparator. The reference voltage is
applied to the inverting input terminal in the comparator circuit. The
IR receiver is attached to the non-inverting input terminal.
The Author Pragadeesh said that it is a reflecting sensor that
suppresses visible light by combining an infrared emitter with a
phototransistor in a lead package.
One of the most important requirements for optimal performance is
that the IR transmitter and receiver be in a straight line. The
transmitter sends infrared rays into the driver's eye.
There will be a high output for closed eyes and a low output for open
eyes depending on whether the eye is closed or open. The IR receiver
picks up the transmitted signal. The comparator is attached to this
receiver.
ADVANTAGES
● The component quickly makes a connection with other drivers.
● The driver's life can be preserved if the vehicle is locked by
ignition system.
● By limiting the number of vehicles on the road, traffic control
can be maintained.
● Accidents and traffic congestion are preventable.
FUTURE SCOPE
The number of consecutive frames where the eyes are closed is all
that matters to this system. It may be too late at that moment to
provide a warning.
By researching It is feasible to find a pattern of eye movement. a way
for generating the warning more quickly Another option for locating
information is to use 3D photos.
The pupils of the eyes The eyes are the most detailed feature of a 3D
model
This could be a more durable method off occusing on the eyes We can
employ Automatic Braking instead of an alarm.
A system that reduces the vehicle's speed. We may park the car
autonomously by first using Automatic braking device that slows
down the vehicle. The car will start and the parking lights will switch
on at the same time car's lights and will detect the parking place and
will park the car autonomously, saving an accident.
CONCLUSION
● Eye-based control will be the future of all device
control, making operations far more pleasant and
simple with minimal human intervention.
● Several high-risk operations are possible be easily
accomplished with this type of application.
● Research and study in these areas will usher in a new
era of innovation communicating with computers.
● As a result, a mechanism to track fatigue has been
developed. It was created by detecting eye blinks and
head movement.
● With a 94 percent accuracy rate and a 1% false
positive rate, the suggested system recognises eye
blinks. Our tests revealed that the proposed approach
provides quick and accurate results.
● Sleepiness detection is a term used to describe the
detection of drowsiness. According to the facts, The
most important factors that influence the outcome of
experiments are:
● The presence of glasses and the effectiveness of our
method a lot of lighting changes in the presence of
glasses is undeniable.
● It have an impact on the system's essential
components, such as the face detection, detection of
the eyes, and symmetry calculation
REFERENCE
[1] C. Liu and R. Subramanian, “Factors Related to Fatal Single-Vehicle
Run-Off-Road Crashes,” U.S. Department of Transportation, American
National Highway Traffic Safety Administration, DOT HS 811 232,
Washington, D.C., November, 2009.
[2] R.R. Knipling and J.S. Wang, “Crashes and Fatalities Related to Driver
Drowsiness, Fatigue,” National Highway Traffic Safety Administration,
Washington, D.C., Information Management Consultants Inc., McLean,
VA., 1994.
[3] G. Pan, L. Sun, Z. Wu and S. Lao, “Eyeblink-based Anti-spoofing in
Face Recognition from a Generic Webcamera,” The 11th IEEE
International Conference on Computer Vision (ICCV'07), Rio de
Janeiro, Brazil, October, 2007.
[4] J. Santamaria and K. Chiappa, “The EEG of drowsiness in normal
adults,” Journal of Clinical Neurophysiology, vol. 4, no. 4, pp. 327-382,
1987.
[5] R.P. Nikhil, C. Chien-Yao, L.W. Ko, C.F. Chao, T.P. Jung, S.F. Liang
and C.T. Lin, “EEG-based Subject-and Session-independent Drowsiness
Detection: An Unsupervised Approach,” EURASIP Journal on Advances
in Signal Processing, ISSN: 1110-8657, January 2008.
[6] A. Vuckovic, V. Radivojevic, A.C.N. Chen and D. Popovic, “Automatic
recognition of alertness and drowsiness from EEG by an artificial neural
network,” Medical Engineering & Physics, vol. 24, no. 5, pp. 349, 2002.
[7] B. James, H. Sharabaty and D. Esteve, “Automatic EOG analysis: A first
step toward automatic drowsiness scoring during wake-sleep
transitions,” Somnologie, vol. 12, pp. 227-232, 2008.
[8] J.H. Yang, Z.H. Mao, L. Tijerina, T. Pilutti, J.F. Coughlin and E. Feron,
“Detection of driver fatigue caused by sleep deprivation,” IEEE
Transactions on Systems, Man and Cybernetics, vol. 39, July 2009.
[9] C.C. Liu, S.G. Hosking and M.G. Lenné, “Predicting driver drowsiness
using vehicle measures: Recent insights and future challenges,” Journal
of Safety Research ,vol. 40, no. 4, pp. 239-245, August 2009.
[10] D. Dinges, “PERCLOS: A valid psychophysiological measure of
alertness as assessed by psychomotor vigilance indianapolis,” Federal Highway
Administration, Office of Motor Carriers, Technical report,
MCRT-98-006, 1998.
[11] W.W. Wierwille, L.A. Ellsworth, S.S. Wreggit, R.J. Fairbanks and C.L.
Kirn, “Research on vehicle based driver status/performance monitoring:
development, validation and refinement of algorithms for detection of
driver drowsiness,” National Highway Traffic Safety Administration,
Technical report, DOT HS 808 247, 1994.
[12] F.M. Sukno, S.K. Pavani, C. Butakoffand and A.F. Frangi, “Automatic
Assessment of Eye Blinking Patterns through Statistical Shape Models,”
ICVS 2009, LNCS 5815, Springer-Verlag Berlin Heidelberg, pp. 33-42,
2009.
[13] P.P. Caffier, U. Erdmann and P. Ullsperger, “The spontaneous eye-blink
as sleepiness indicator in patients with obstructive sleep apnoea
syndrome - a pilot study,” Sleep Medicine, vol 6., no. 2, pp. 155-162,
2005.
[14] T. Morris, P. Blenkhorn and F. Zaidi, “Blink detection for real-time eye
tracking,” Journal of Network and Computer Applications, vol. 25, no.
2, pp. 129-143, April 2002.
[15] M. Divjak and H. Bischof, “Eye blink based fatigue detection for
prevention of Computer Vision Syndrome,” MVA2009 IAPR
Conference on Machine Vision Applications, Yokohama, Japan, May
2009.
[16] M. Chau and M. Betke, “Real Time Eye Tracking and Blink Detection
with USB Cameras,” Boston University Computer Science Technical
Report No. 2005-12, 2005.
[17] L. Wang, X. Ding, C. Fang and C. Liu, “Eye blink detection based on
eye contour extraction,” Proceedings of SPIE, vol. 7245, 72450R, 2009.
[18] P. Viola and M. Jones, “Rapid object detection using a boosted cascade
of simple features,” International Conference on Computer Vision and
Pattern Recognition, CVPR 2001.
[19] H.A. Rowley, S. Baluja and T. Kanade, “Neural Network-Based Face
Detection,” IEEE Transactions on Pattern Analysis and Machine
Intelligence, vol. 20, pp. 23-38, [Link]
1998.
[20] S. Milborrow and F. Nicolls, “Locating facial features with an extended
active shape model,” In D. Forsyth, P. Torr and A. Zisserman (eds.)
ECCV 2008, Part IV. LNCS, vol. 5305, 504--513. Springer-Verlag,
Heidelberg, 2008.
[21] T.F. Cootes, G.J. Edwards and C.J. Taylor, “Active appearance models,”
In [Link] and B. Neumann, editors, 5th European Conference on
Computer Vision, vol. 2, 484--498. Springer, Berlin, 1998.
[22] A. Ortega, F. Sukno, E. Lleida, A. Frangi, A. Miguel, L. Buera and E.
Zacur, “AV@CAR: A spanish multichannel multimodal corpus for
in-vehicle automatic audio-visual speech recognition,” In M.T. Lino,
M.F. Xavier, F. Ferreira, R. Costa and R. Silvia, editors, IV International
Conference on Language Resources and Evaluation, vol. 3, pp. 763-767,
European Language Resources Association, May 2004.
[23] E. Vural, M. Cetin, G. Littlewort, M. Bartlett and J. Movellan “Drowsy
Driver Detection Through Facial Movement Analysis,” Lecture Notes in
Computer Science, vol. 4796, pp. 6-18, 2007.
