Classification of Blink Type by a Frame Splitting Method using Hi-Vision Image

Abstract

Recently, the human-machine interface using information of user's eye blink was reported. This system operates a personal computer in real-time. Eye blinks can be classified into voluntary (conscious) blinks and involuntary (unconscious) blinks. If the voluntary blinks can be distinguished in automatic, an input decision can be made when user's voluntary blinks occur. By using this system, the usability of input is increased. We have proposed a new eye blink detection method that uses a Hi-Vision video camera. This method utilizes split interlace images of the eye. These split images are odd- and even- field images in the 1080i Hi-Vision format and are generated from interlaced images. The proposed method yields a time resolution that is double that in the 1080i Hi-Vision format. We refer to this approach as a “frame-splitting method”. We also proposed a method for automatic eye blink extraction using this method. The extraction method is capable of classifying the start and end points of eye blinks. In other words, the feature parameters of voluntary and involuntary blinks can be measured by this extraction method. In this paper, we propose a new classification method for eye blink types using these feature parameters.