This document describes the design and implementation of an eye movement controlled wheelchair. It uses a webcam to capture eye movements which are then processed using MATLAB to detect the direction of movement. Based on the detected movement, MATLAB sends a signal over a serial port to a microcontroller which controls the motors to move the wheelchair in the intended direction. The system was tested for accuracy and safety features were incorporated like speed control and halt functions. While the performance was around 70-90% accuracy, further work is needed to improve reliability for commercial use.
3. INTRODUCTION:
Many people are suffering from
quadriplegia around the world,
which affects their four limbs.
So we made a project called eye
monitored wheel chair
movement which makes their life
easy
We take the advantage of their
ability to move their head.
We use laptab webcam to
capture their head movements
by that the wheel chair can be
moved to desired direction.
4. DESIGN OVERVIEW:
• Eye detection can be done by MATLAB script which is running on the
laptab which has inbuilt webcam.
• MATLAB puts one digit like 0,1,2 on COM port based on the eye
movements
• ATMega 32A is used to drive motors as per user desire direction like
left,right,straight based on signals send by matlab
5. SOFTWARE DESIGN:
(A) MATLAB COMPONENT
Initialization of COM port for serial communication.
Capturing of image for every 25th frame.
Changing image from RGB to gray level and enhancing the
image.
Detection of the person’s eye in image using viola-jones
algorithm.
Detection of the eye movement by comparing the present
position of eye with previous position.
Placing one digit data on the COM port based on the eye
movements.
6. Code developed in AVR studio.
Hex file generated using AVR
studio.
Hex file dumped into
Microcontroller using extreme
burner software
(B) FIRMWARE DESIGN
7. HARDWARE DESCRIPTION
(A)DESCRIPTION OF MICRO
CONTROLLER ATMEGA
32A
• ATMega 32A is a 8-bit
microcontroller having 40
pins
• ATMega 32A has a 32
Programmable I/O Lines
• ATMega 32A has a
programmable USART
• ATMega 32A can operate
over 10,000 Flash/100,000
EEPROM Write/Erase
cycles.
8. MECHANICAL DESIGN
WHEEL CHAIR PROTO TYPE
Metal Chassis – It is flat
metal body which
resembles wheel chair
DC motors – Two DC
motors are used having
200 RPM
Wheels – Two types of
wheels. One is castor
wheel and other is plastic
wheel(two wheels)
11. CP2102
CP2102 is a interfacing device between computer and
microcontroller
CP2102 carry one digit data put on COM port by
MATLAB using RS-232 standard
It also converts RS-232 voltage levels into TTL voltage
levels
CP2102 has RXD,TXD,DTR,VBUS,GND,3V3 pins
12. RS-232 STANDARD
RS-232 standard has 1 start bit,8 data bits, 1 parity bit and 1 or 2 stop bits
RS-232 has +12v volts for logic 1 and -12v for logic 0
13. (B) MICROCONTROLLER – DC MOTORS
INTERFACE
To interface MCU with
motors we use L293D IC.
Based on the input data
output pins are enabled.
Dc motors rotate based
on the logic levels at
IN1,IN2,IN3,IN4 of L293D
16. Testing strategy – x, y, length, height
of valid eye are displayed on the
debug screen.
Speed of execution – This is the
limitation of our project. We can take
image for every 15th frame but cannot
process at that speed.
Accuracy – can get almost 86%
accuracy with coordinates of eye that
are set in threshold limits
17. SAFETY FEATURES
Controlled speed of detection and wheel chair drive.
Eye height and width threshold.
Controlled movement in either direction for limited time period.
Less jerk, by incorporating delay in the wheel chair drive.
Halt control using blink detection.
18. CONCLUSION
PERFORMANCE
System functions at an accuracy of 70 – 90%.
Person’s eye can easily detected by using viola- jones algorithm without
detecting of things mostly have face like features.
FUTURE MODIFICATIONS
• Lot of work needs to be done before making the product commercially viable.
• Need to work on wheel chair movement when the person is just casually glaring
in different directions.
• We can incorporate certain sequences for turning ON and OFF electrical devices
or door locks.
• Lot of safety precautions need to be incorporated.
19. REFERENCES:
1. S. Tameemsultana and N. Kali Saranya, “Implementation of Head and Finger
Movement Based Automatic Wheel Chair”, Bonfring International Journal of Power Systems
and Integrated Circuits, vol. 1, Special Issue, pp 48-51, December 2011.
2. Manuel Mazo, Francisco J. Rodriguez, Jose L, Lazaro, Jesus Urena, Juan C. Garcia,
Enrique Santiso, Pedro Revenga and J. Jesus Garcia, “Wheelchair for Physically Disabled
People with Voice, Ultrasonic and Infrared Sensor Control “, Autonomous Robots, vol.2, no.
3, pp. 203-224 ,Sep 1995.
3. Tabasum Shaikh, Naseem Farheen Sayyed, Shaheen Pathan, “Review of Multilevel
Controlled Wheelchair”, 4th National Conference On Electronic Technologies, pp. 275-
279, April 2013.
4. Motor Controller circuitry –
http:// en.wikipedia.org/wiki/Motor controller
5. Voila Jones Algorithm -
http://en.wikipedia.org/wiki/Viola%E2%80%93Jones_object_detection_framework
6. Cascade Object Detector –
http://www.mathworks.com/help/vision/ref/vision.cascadeobjectdetector-class.html