Abstract – In this world of technology everything is being computerized. But there are many people with several disabilities like handicapped people who are unable to take advantage of this technology. This system proposed an efficient approach to handle the computer. The first approach is to use the webcam to control the cursor, the second approach researches speech recognition as an input and the third approach deals with the Fusion of both Eye and Speech. As most of the new Computer System comes with built in Webcams and Microphones, it will be very beneficial and cost effective.
Keywords- Face Capturing, Image Extraction, Harris Corner Detector, Contours, Cursor Controlling, Speech Recognition.
Computer is most common device which is used by individuals all over the world. It is the most intelligent machine available for the computational purpose and lots of input output devices are developed to interact with it (like mouse, keyboard, joystick, scanner, printer and many other).The main motivation behind designing every new system is to make interactions in more humanly fashion. Every new device can be seen as an attempt to make the computer more intelligent and making humans able to perform more complicated communication with the computer.
In the world of humans most of the communications are done by using vision and sound. So this interface will prove better than any other available systems. The most important factor using this system is that the user does not have to be physically in contact with contact with the system. And if we consider noisy environment it will prove better than that.
In this paper we are proposing feature based detection and image localization algorithms , , . The advantage using this algorithm is that it performs faster but it requires high quality images. Rather we can use combined algorithms such as neural network, genetic algorithm , ,  which makes the use of low quality images. Here we are considering steady state fixes system which will continuously detect head movements of the user and capture images of eye. The eye features are extracted in real time using the images captured .From extracted eye features eye gaze is detected to control the cursor. It can be divided into two forms viz. Head mounted , ,  and steady state fixed. Both will have their advantages and disadvantage depending on the environment they are used.
A. Human Eye Structure
The eye is not shaped like a perfect sphere; rather it is fused two- piece unit. The smaller frontal unit, more curved called the Cornea is linked to the larger unit called SCLERA black centre, the pupil are seen instead of the cornea due to cornea’s transparency. The cornea segment is typically about 8 mm in radius. The sclerotic chamber constitutes the remaining 5/6, its radius is about 12 mm. The cornea and sclera are connected by a ring called limbus.
The eye is made up of three coats, enclosing three transparent structures. The outermost layer, known as fibrous tunic, is composed of cornea and sclera. The middle layer, known as the vascular tunic or uveas, consists of the choroid, cilliary body, and iris. The innermost is the retina, which gets its circulation from the vessels of the choroid as well as the retina vessels which can be seen in an ophthalmoscope.
B. Corner Detection
1. What Are Corners’? Intuitively, junctions of contours.
‘ Generally more stable features over changes of viewpoint.
‘ Intuitively, large variations in the neighborhood of the point in all direction.
‘ They are good features to match.
Fig 1: Corners (Junction of Contours)
2. Corner Points: Basic Idea
‘ We should easily recognize the point by looking at intensity values within a small window
‘ Shifting the window in any direction should yield a large change in appearance.
Fig 2: Detection of Corner Points.
3. Harris Corner Detector: Basic Idea
The most common idea of corner is defined by Harris . The Harris corner detector  is based on local auto-correlation function of signal.
Fig 3: Mathematical approach for determining which case holds.
4. Harris Detector: Mathematics
...(download the rest of the essay above)