Video Oculographic System using Real-Time Video Processing

Abstract

Video Oculography (VOG) is a non-invasive, video based method of measuring horizontal, vertical and torsion position components of movements of both eyes using a head mounted mask that is equipped with small cameras. There are systems which use various techniques to detect the opening and closing of the eyes. The technique used in our project is simple and not time consuming. Using blinks to select a sentence or word and playing the sentence has not been used widely in the past. In our project we want to implement this and also build a system whose functionality can be extended to serve a variety of applications. In this system, we use a camera to take input from the user. The user blinks in front of the camera. This is taken as a video sequence and then it becomes the input to blink detection and counting code. The number of blinks is thus calculated and the sentence corresponding to this count is played. This system works with minimum specification. This would help the lower economical class. There are several methods which can be used to detect the eyes, the method which we have used in our project reduces the time taken for blink detection and also reduces the cost of the system by a large margin.

Eye Blink Detection System for Paralyzed Patients

“…Eye region in the image was extracted by segmentation technique. The recorded movement of eye is converted to speech [5]. Jacek et.al , has developed a VOG system based on a web camera which captures minute information about head shaking and eye position of a Parkinson's disease patients.…”