Robust and Low-Computational Cost Eye Tracking System Using a Generic Webcam

Eye-tracking is a process for measuring eye movement and gaze location. It's considered a standard approach for inspecting the visual attention allocation. In this paper, a new eye-tracking approach was tested and evaluated principally for the first time. The detected gaze location values were used to control the mouse cursor in real time on a computer device. A generic web camera was placed on the eyeglasses frame and aimed at the eye; no Infrared light source was used to illuminate the eye. Every captured video frame of the eye is transformed to a grey level, then the contrast is enhanced using histogram equalization. Finally, the image of the eye is identically divided into four groups. The eye gaze position in the horizontal direction is detected by calculating the difference between the left and right groups of pixels. Similarly, the eye gaze in the vertical direction is detected by calculating the difference between lower and upper groups of pixels. The results show high accuracy, low computational cost, and confirmed perfect linearity. Moreover, it is a non-invasive method, and easy to calibrate.