Development of head-mounted eye tracking system achieving environmental recognition ability

This paper presents development of a new head-mounted eye tracking system for human adaptive mechatronics (HAM) system. Estimation of human skill level is important to assist human suitability. Eye movement indicates human insight such as changing attention and intention, hence this paper focuses on design of a headgear to reduce operator´s blind side and to extend scene camera image which superimposes point of gaze. By using a robust pupil locator based entirely on integer calculations for easy real-time implementation in FPGA hardware, an algorithm estimating pupil´s center against ill-conditional images contaminated by eyelid and eyelash is proposed.