In this modern era of science and technology, several innovations exist for the benefit of the differently-abled and the diseased. Research organizations worldwide, are striving hard in identifying novel methods to assist this group of the society to converse freely, move around and also enjoy those benefits which others do. In this paper, we concentrate on assisting people suffering from one such deadly disease – the Motor Neuron Disease (MND), wherein a patient loses control of his/her complete mobility and is capable of only oculographic movements. By utilizing these oculographic movements, commonly known as the eye-gaze of an individual, several day to day activities can be controlled just by the motion of the eyes. This paper discusses a novel and cost effective setup to capture the eye gaze of an individual. The paper also elaborates a new methodology to identify the eye gaze utilizing the scleral properties of the eye and is also immune to variations in background and head-tilt. All algorithms were designed on the MATLAB 2011b platform and an overall accuracy of 95% was achieved for trials conducted over a large test case set for various eye gazes in different directions. Also, a comparison with the popular Viola-Jones method shows that the algorithm presented in this paper is more than 3.8 times faster.
All Science Journal Classification (ASJC) codes
- Control and Systems Engineering
- General Computer Science