On an eye-gaze input system based on the limbus tracking method by image analysis

摘要

Eye-gaze input devices as advanced human-machine interfaces have already been developed. They are often used as communication aid for the seriously physically handicapped like ALS patients. The conventional eye-gaze input systems can, however, harm the user. This is due to the use of an infrared ray on the eyes to detect eye-gaze. It is said that the prolonged irradiation could cause eye diseases such as glassblower's cataract, retinitis solaris and dry eyes. We propose a new eye-gaze input system using a personal computer and a home video camera for eye gaze detection under natural light. Our usual system has horizontal 10 indicators and the user can select any indicators by eye-gaze of user. We have been also developing the method of vertical eye-gaze detection with compensation of user's head movement. This method can distinguish vertical 3 indicators.