EYE GAZE AND INFORMATION SECURITY USING WEBCAM Submitted in partial fulfillment of the Requirement of the course of Management Information System Bachelor of Science
EYE GAZE AND INFORMATION SECURITY USING WEBCAM
Submitted in partial fulfillment of the
Requirement of the course of
Management Information System
Bachelor of Science (Computer Science)
Bahria University Karachi Campus
Sikander Sial Khan
Syed Hamza Ali Rizvi
Eye gaze is natural form of interaction, accomplished by identifying where a person is looking or in easier words where is the focus of human’s eye. The technology makes possible the development of new human machine interfaces, the act of looking a screen is part of most natural interaction processes but the information that the eye gaze can give us is still not entirely exploited today, gathering and processing users eye gaze to interact with machine is a topic already studied but mostly is based on specific technologies that are not available in mass market devices, such as laptops, tablets or ATM machine. This work describes a system to detect eye gaze based on a laptop web camera enabling a more natural form of human machine interaction.
2. Background and Related Work
2.1. Human Eye
2.1.1. Eye Structure
2.1.2. Eye Blinking
2.2. Methods of Finding Eye Blink
2.2.2. Haars Classifiers
220.127.116.11. Voila Jones
3.1. Web camera
3.2. Programming language
3.3. General Algorithm
3.3.1. Loop Steps
3.3.2. Main Program
4. Benefits of Humans Eye interaction
KEYWORDS: Human Computer Interaction, Face Detection, Eye Blinking, Eye Structure, Haar-like Features, Voila Jones, OpenCV
Human Computer interaction (HCI) is a way to communication with Computer Machine. Commonly used input devices include the following: keyboard, computer mouse, trackball, touchpad, touch-screen and many more. Eye gaze is a natural form of interaction, accomplished by identifying where a person is looking. The advance technology makes possible the development of new human- machine interfaces. The act of looking a screen is part of most natural interaction processes. But the information that the eye gaze can give us is still not entirely exploited today. Gathering and processing users’ eye gaze to interact with machine is a topic already studied, but mostly is based on specific technologies that are not available in mass market devices, such as laptops, tablets or ATM machine. This work describes a system to detect eye gaze based on a laptop web camera enabling a more natural form of human-machine interaction.
So this form of interaction we can develop industry product such as information security using eye gaze through webcam with simple eye blink.
2. BACKGROUND AND RELATED WORK
Eye gaze is a natural form of interaction, accomplished by identifying where a person is looking or where is the person’s eye focused. However, replicating this procedure automatically within the scope of human-machine interaction is not simple. Eye gaze techniques are being studied for more than hundred years and has remained a subject of several studies over the last decades 1,2,3. The first study that was unveiled was performed by Rayner and Pollatsek. They developed a system based on electro-oculography using electrodes on human skin that measured electric potential differences. Duchowski developed a system based on contact lenses. However, most reliable systems which produce the best results use specific and expensive equipment. New system are being developed that use non-intrusive and more affordable devices. Among these systems, the best performances are obtained with a source of infrared light. Distrust of infrared light exposure motivated the development of eye gaze detection systems that use current technology, such as webcams. These system still have limitations associated with head movement’s compensation and eye random and involuntary movements. It is also necessary to improve real time processing algorithms and hardware.
There is a commercial application developed in Portugal by Luis Figuiredo which implements this kind of human-machine interface. However, it uses a higher level and expensive hardware, infrared and high speed camera which increases the commercial cost. Because of expensive and harmful infrared rays for eyes this interface fails using infrared rays.
Gaze controlled and eye-blink-controlled user interfaces belong to the second group of systems suitable for the people who cannot speak or use their hands to communicate. Most of the existing methods for gaze communication are intrusive or use specialized hardware, such as infrared (IR) illumination devices or electrooculography (EOG). Such systems use two kinds of input signals: scan path (Line of gaze determined by fixations of the eyes) or eye-blinks. The eye-blink-controlled systems distinguish between voluntary and involuntary blinks and interpret single voluntary blinks or their sequences. Particular eye-blink patterns have the specific keyboard or mouse commands assigned, e.g., a single long blink is associated with the TAB action, while a double short blink is a mouse click. Such strategies can be used as controls for simple games or for operating programs for spelling words.
A more efficient system was described in. It uses two Webcams—one for pupil tracking and second for estimating head position relative to the screen. Infrared markers placed on the monitor enable accurate gaze tracking. The developed system can replace the computer mouse or keyboard for persons with motor impairments. The active approach to eye and eye-blink detection gives very accurate results, and the method is robust. The advantages of the IR-based eye-controlled human–computer interfaces are counterbalanced by high end-user costs due to specialized hardware.
2.1. HUMAN EYE
The Human eye is an organ which reacts to light and pressure. As a sense organ. Human Eyes help to provide a three dimensional, moving image, normally colored in daylight. To understanding the working of the eye gaze tracking. First to understand the relevant biology and behavior of the human eye.
2.1.1. EYE STRUCTURE
Our Eye has almost an oval-shaped structure. It is not a proper oval and eye is slightly asymmetrical globe, about an inch in diameter. We have many thing present in our eyes i.e. cornea, iris, pupil, sclera, conjunctiva etc.
2.1.2. EYE BLINKING
There are two types of eye blink as followed:
• Voluntary Blink (unintentionally blink)
• One eye blink or Involuntary blink (intentionally blink)
2.2. METHODS OF FINDING EYES BLINKS
The algorithm used by the system for detecting and analysing blinks is initialized automatically by web cam, dependent only upon the inevitability of the involuntary blinking of the user. After detecting the blink pattern of blinking eyes go through the pattern recognition process after that the system will perform upon the nature of eye blink e.g: Left eye blink or right eye blink.
Eye Gaze system uses several different techniques of which eye tracking is only a part. There are many great methods which are created through combination of many techniques.
(1) Face detection
(2) Eye-region extraction
(3) Eye-blink detection
(4) Eye-blink classification.
2.2.1. HAARS CLASSIFIERS
Haar Classifier objects are based on compilation of Haar-Like features. Those features use the changes in contrast values between adjacent rectangle of pixels to determine relative light and dark areas. Two or three rectangles with relative contrast difference from single Haar-like feature. These features, as shown in figure 18.104.22.168 below, are used to detect object in an image. These features can be scaled up and down easily by increasing or decreasing the size of the pixel group. This allows Haar-like features to detect objects of various sizes with relative case.
22.214.171.124. VOILA JONES
The Viola jones object detection framework is the first object detection framework to provide competitive object detection rates in real-time. Although it can be trained to detect a variety of object classed. i.e. face detection, eye detection, pattern detection, blink detection etc.
The problem is that the human can easily identify the face, eye, any pattern but computer needs to precise instructions and constraints. To make the task more manageable.
This algorithm/framework has four stages:
• Haar Feature selection
• Creating an integral image
• Adaboost Training
• Cascading Classifiers
In order detect the user’s eye blink we need to know the voluntary blink and involuntary blink and other features we can work out the where the gaze is. The system does not need the powerful or expensive camera for the blink detection.
There are a large number of algorithm available for use, which deploy an even larger number of varying techniques. The fundamental requirements appear to be face detection eye detection, blink detection. As such, this will be how this system will be designed.
The aim of this research paper is to create a Gaze tracking system using simple techniques, a regular webcam and Computer Vision library OpenCV in C#. Since the logical design the most Gaze Tracking systems is fairly standard, the overall design of this system is very similar to existing systems. However, the tools used in these systems have a large amount of variation in them. As such, the particular techniques used were chosen to be simple techniques that were easily implemented as a baseline comparison.
This system will perform on real time using eye blink detection as the basis of interaction commands.
3.1. WEB CAMERA
The web camera is use for this project is simple laptop or any other external web camera of 1.3 megapixel capabilities. This is a relatively cheap web camera and as such suited the purposes of the project.
3.2. PROGRAMMING LANGUAGE
A simple choice to make is which programming language the system should be implemented in. OpenCV supports three language as default: C#, C++ and Python.
A simple test was then run on python to see if it was possible to attempt to keep the system in real time but python was found to be insufficient, C++ and C# was the next choice to simplify the implementation of the system. This is because C++ and C# has support in Visual Studio and has been covered in the undergraduate curriculum.
3.3. GENERAL ALGORITHM
The general algorithm is split in two different parts
• Loop Steps
• Main Program
3.3.1. LOOP STEPS
The loop steps required to get all the necessary data from a single frame.
Input ? Face Detection ? Eye Detection
? Blink Detection ? Output
First step the camera should input the image frame and then system detect the face after detection of face system process face and identify eyes after eyes the focusing on set the pixels near the eyes after that detection of blink process if blink detect properly then task perform in the form of output.
3.3.2. MAIN PROGRAM
The main program incorporates the loop steps to create the gaze tracking system, initially the calibration program is run to get the face and eye position as shown in the figure 126.96.36.199 below:
The system will go through the calibration stage first, in order to provide reference points for the main detection, after which the main program begins running a continuous loop that calculates the detection of eye and blink after detection show some output.
4. BENEFITS OF HUMANS EYE INTERACTION
The benefits of the human eye interaction are innumerable. Controlling electronic gadgets with our eyes in a cheaper way. After a successful detection of eye, it is just the work of programming for what to do with eye’s blink detection, thus performing of almost every task in for example, PC, from opening it up to shut it down.
As far as security is concerned it is required in almost every field of life. Technology has emerged as a part of life thus the security is required in that case too. To fulfill the need of security, many types of security software and hardware are introduced. Eye gaze is a new addition in it as it totally depends upon the movement of one’s eye. Unlike fingerprint scanning it will be very much time saving and revolutionary in almost every field of life. The eye-detection unlike other security measures, is very secure as it will almost be unable to hack, because human’s brain cannot be hacked by the hacker, and it is function of the human brain to set the blinks to be detect by the PC to perform particular task.
The eye-gaze technique is going through massive research. And in the result of that research, many eye-gaze techniques have came out. But the discoveries made until today lacks accuracy and easier market availability. The problem is that eye gaze techniques discovered till today are not very much accurate. We want to evade this problem by using cheaper and easier way of solving it. The purpose of this research paper is that the world needs more accurate and more cheaper way to interact and perform tasks by using human eye. The laptop’s camera (or webcam in case of desktop PC) to detect human eye and then perform particular operations is the easiest, quickest and cheapest way to perform such tasks. We will get through this problem by simply detecting human’s eye. After the successful detection of human’s eye, we will be able to perform many tasks.
1 Yukiko I.Nakano, Cristina Conati, Thomas Bader, Eye gaze in Intelligent user interface, Springer London Heiderlberg New York Dordrecht
2 Michael Chau and Margri Betke, “Real Time Eye Tracking and Blinking Detection” Boston University, Boston, MA 02215, USA. May-12 2005
Technical Reports and Theses:
3 Drewes, Heiko (2010): Eye Gaze Tracking for Human Computer Interaction. Dissertation,LMU München: Faculty of Mathematics, Computer Science and Statistics 18. March 2010
Electronic Sources from Internet:
4 Drewes, Heiko Eye Gaze Tracking for Human Computer Interaction. Dissertation, LMU München: Faculty of Mathematics, Computer Science and Statistics and 2010. On Line, Available at: https://edoc.ub.uni-muenchen.de/11591/