The method of eye tracking in E-learning figure detection

Students and the teacher can´t make face to face communication in the condition of E-learning. In this situation, the teacher can´t know the status of students and can´t take corresponding measures to adjust the students. A method was proposed to detect and track eyes and to recognize the opening and shutting of eyes in a real-time video stream. The eye was tracked by detecting the inner eye corner. Open eye template could be created online and dynamically updated. The opening or shutting of eyes could be recognized using correlation with an online open eye template. Experiments indicate that an accuracy rate of 95% in blink detection is obtained using the proposed method at a processing rate of 50 fields per second, which could be applied to E-learning figure Detection successfully.