フレーム分割法による瞬目の自動抽出

概要

論文の詳細を見る
We have proposed a new eye blink detection method that uses a home video camera. This method utilizes split interlace images of the eye. These split images are odd- and even- field images in the NTSC format and are generated from NTSC frames (interlaced images). This proposed method yields a time resolution that is double that in the NTSC format. That is, the detailed temporal change occurring during the process of eye blinking can be detected. However, this method yields vertical space information that is half that in full sized NTSC images. To improve this point, we apply a hi-vision camera to measure eye blink by our proposed method. In the result, the dynamic range of eye blink detection increases. Further, we propose a new method to extract the wave pattern at eye blinking. By using this method, the eye blink can be extracted from the detection result automatically. The extraction method classifies the wave pattern at eye blinking by using the statistical information of eye blinking wave pattern.
2012-09-01