This Presentation aims at explaining how eye tracking works and the usage of Houghman Circle Detection Algorithm in order to detect the iris.
https://www.picostica.com
2. Introduction
• An eye has a lot of communicative power. Eye contact
and gaze direction are central and very important
factors in human communication.
• The Computer mouse has gained high popularity as an
input device.
• The Eye Tracking System is a Human Computer
Interaction application where a person can control
the mouse pointer using his/her eyes.
• Useful for Paralyzed / Physically disabled people to
convey their messages to others.
3. • Eye Tracking is the process of
measuring where we look, also
known as our Point of Gaze.
• Eye Tracking is a technology that
puts you in control of your device by
using your eyes as you naturally
would.
• A device or computer equipped with
eye tracker knows what a user is
looking at. This makes it possible for
users to interact with computers
using their eyes.
4. How Human Eye Works?
Light enters the eye through the cornea.
From the cornea, the light passes through the pupil. The amount of
light passing through is regulated by the iris, or the colored part of
the eye.
From there, the light then hits the lens, the transparent structure
inside the eye, which focuses light rays onto the retina.
Finally, it reaches the retina, the light-sensitive nerve layer that lines
the back of the eye, where the image appears inverted.
The optic nerve carries signals of light, dark, and colors to the area
of the brain, which assembles signals into images (our vision).
5. Eye Tracking Terminology
Fixation – A single instance where the eye momentarily stops
Saccade - The rapid eye movement from one fixation to the next
Scan Path – A series of sequential fixations and saccades.
6. Goal of the System
• Hands – free Computing
• Facilitating the handicapped in using the Computer
• Controlling the mouse pointer through eye movement.
• Eye based Human – Computer interaction provides real Time Eye Tracking
and Eye - Gaze estimation.
7. • Image Acquisition: A web – cam is required to acquire images. System will
start with image acquisition using an integrated web-cam or USB web-cam.
• Image Pre-processing: Pre-processing the data is always helpful in
extracting useful information. Images acquired by web-cam must be
converted into appropriate formats required by different algorithm at next
stages.
• Eyes Features Detection: Necessary eye features are identified, Using HCT
for pupil detection and Template Matching Technique for eye corners
detection.
• Screen Mapping: Mapping the coordinates of the screen with the pupil
movements is done.
8.
9. System Components and interactions of the project virtual mouse system.
• The software application is organized into two parts:
The eye tracking engine part and mouse simulation part.
The eye tracking engine part matches our software application with eye tracking
device to eye control mouse function.
10. MODULE DESCRIPTION
Main Interface Module : It Sets and manages the startup dialogue.
Mouse/Keyboard Engine Simulation Module : It create dynamic tool window, automatically
changes the size of the tool window according to the screen resolution, and decides whether to
display or obscure the toolbar.
User Action Detection Module : The fixation function in this module is defined to calculate the
gaze time of the user and determine the mouse coordinates.
Halt Module : It determines whether to stop eye tracking and enter the sleep mode and
determines whether to jump out of the sleep mode and restart the eye tracking.
Mouse Function Module : It receives the view point of user from simulation engine and transfers
the coordinate to the fixation function within user action detection module.
11. Algorithm Description
• Eye Detection using Viola - Jones Algorithm
• Detecting Iris using Circular Hough Transform Algorithm
12. Viola – Jones Algorithm
Computers cannot detect face very easily as humans.
When an image is prompted to the computer, all that it sees is a matrix of numbers.
A regular image is composed of thousands of pixels. Even a small 28 X 28 image is composed by
784 pixels.
Each pixel can assume 255 values. So that’s 255784 possible values.
Viola Jones algorithm extracts a much simpler representations of the image, and combine those
simple representations into more high – level representation in a hierarchical way.
13. STEPS INVOLVED IN VIOLA – JONES ALGORITHM
Haar – like Feature Extraction
We have some primitive masks. Those masks are slide
over the image, and the sum of the values of the pixels
within the white side is subtracted from the black side.
The result is a feature that represents that region.
14. Circular Hough
Transform
Algorithm
Now we detected the eyes using the viola – jones algorithm, the next
step is to detect the iris.
For that we are going to look for the most circular object in the eye
region.
15. STEPS INVOLVED IN CIRCULAR HOUGH
TRANSFORM
• By using CHT, the circle parameters within a given image are
identified: the circle centre coordinates “a” and “b” and the
circle radius “r”, as represented in the circle equation:
• To improve precision we implement a pre-processing step:
– Segmentation threshold to cut image components such as
eyelashes and eyebrows and keep only the pupil contour.
• This is done using the dark pupil technique.
16. HARDWARE AND SOFTWARE REQUIREMENTS
• Operating System – Windows 7 or above, Linux, Mac OS
• Processor – 2Ghz or More
• RAM – 2GB
• Hard Disk – Minimum 120GB
• IR LED
• Web Camera
• OpenCV(Computer Vision)
• MATLAB
• Python IDLE
17. The system combines both the mouse and keyboard functions, so that users can achieve
almost all of the inputs to the computer without traditional input equipment.
The idea of eye control is of great use to not only the future of natural input but more
importantly the handicapped and disabled.
The use of Standard Circular Hough Transformation in image processing module of the
implementation combined with the grid analysis approach is practically successful and has
high potential in future applications.
18. Future Works
One can control home automation system.
Can be used to find the persons interest in a product.
One can control LCD in automobile with the movement of the eye.
Can be used in medical industries while performing high end operations to control
the PC or its various operations.
Can be used by astronauts in space to control various operations inside their suits.