Video Analysis for Detecting Eye Blinking using a High-Speed Camera

Without introducing pre-processing of eye region detection, the eye blinking motion was detected from specific characteristic moving features on a face using a high-speed camera with a frame rate of 250 Hz/sec. In the process of closure and opening during eye blinking, a series of moving points has characteristic features. Both patterns of matching methods and many starting points on a face were examined. Pursuing texture at the bottom of an eyelid results in the perfect detection of eye blinking. Together with behaviors for other non-detectable points, a discriminating power without a false detection was verified.

I. INTRODUCTION

This research started with the aim of developing an automated attention level meter to report whether or not students in a distant classroom with an e-learning system were looking at their teacher on a monitor screen. Eye blinking detection is the first stage for attention level detection, with gaze detection.