A Study of an Intention Communication Assisting System Using Eye Movement

Abstract

In this paper, we propose a new intention communication assisting system that uses eye movement. The proposed method solves the problems associated with a conventional eye gaze input method. A hands-free input method that uses the behavior of the eye, including blinking and line of sight, has been used for assisting the intention communication of people with severe physical disabilities. In particular, a line-of-sight input device that uses eye gazes has been used extensively because of its intuitive operation. In addition, this device can be used by any patient, except those with weak eye. However, the eye gaze method has disadvantages such as a certain level of input time is required for determining the eye gaze input, or it is necessary to present the information for fixation when performing input. In order to solve these problems, we propose a new line-of-sight input method, eye glance input method. Eye glance input can be performed in four directions by detecting reciprocating movement (eye glance) in the oblique direction. Using the proposed method, it is possible to perform rapid environmental control with simple measurements. In addition, we developed an evaluation system using electrooculogram based on the proposed method. The evaluation system experimentally evaluated the input accuracy of 10 subjects. As a result, an average accuracy of approximately 84.82 % was determined, which confirms the effectiveness of the proposed method. In addition, we examined the application of the proposed method to actual intention communication assisting systems.