Eye Controlled Wheelchair with Asthenopia Detection

In this paper we present a low cost and real time system which enables a person with quadriplegia to control his wheelchair with eye movements. This system is enabled by voice control. Once enabled, it processes consecutive frames from the web cam to detect the direction of eye movement. Consequently, a signal is sent to the Raspberry pi micro-controller, to turn the motors in the desired direction. Detection is done by pupil monitoring using contours. Additionally, it detects Asthenopia, a disease caused by straining of the eye. This is achieved by monitoring the blink rate of the patient. Blink Detection is carried out using two methods. The first method uses a combination of SURF and Harris corner detection while second method uses facial landmarks.