Implementation of eye-tracking system based on circular Hough transform algorithm

This paper presents the implementation of an eye-tracking device based on the circular Hough transform algorithm. The proposed device is a part of a complex integrated system for assistance in communicating with and telemonitoring severe neuromotor disabled patients, which consists of an infrared video camera and a PC used to process the captured signals. The eye-tracker device that we have implemented uses keywords technology. It has been tested on a set of subjects by using different parameters for the pupil detection algorithm in order to determine the system optimal performances (algorithm running time and accuracy).