An Image Based Eye Controlled Assistive System for Paralytic Patients

摘要

Communication is an essential part of human life which paralytic patients with locked-in syndrome are deprived of. In locked-in syndrome, the patient cannot move any of his voluntary muscles except the eyes. Taking this into consideration, the proposed system is designed to detect the face and pupil of the patient through a standard webcam using Haar cascade classifiers and Circular Hough Transform algorithm respectively. The proposed system displays different images of daily activities. The patient will have to look at an image for a period above a pre-decided threshold time in order to select it. Subsequently, the system will track the point of gaze of the patient and will select the image accordingly after a confirmation from the patient. Based on this confirmation, the aide will be notified via text or audio. Successful implementation of the system will help the paralytic patient to easily communicate his needs to the aide.