Eye Pupil Based Control of Electric Wheelchair

Abstract

This project is aimed to developed a technique for such disabled people, who suffer with Amyotrophic Lateral Sclerosis (ALS) or quadriplegia and are not able to use a manual or even a motorized wheelchair. As ALS patients or quadriplegic patients are not able to move their hands, or use their hand muscles to perform a basic task, the aim is use such a body part which they are able to use and control properly. Some of these parts are head, tongue and eyes. In this thesis we developed a technique through which a patient will be able to control a wheelchair just by movement of his/her eye. A basic pupil detection technique is developed which is able to detect pupil in almost all kinds of lightning conditions using Infra-Red light. To use the light and dark pupil features of eye, a normal webcam was modified into an infrared camera. An eye gaze detection and estimation method was designed to find the gaze direction of patient. A stop/start and left/right protocol was tabulated and implemented for an easy control of wheelchair using the moving of eye.