鼻孔画像に基づいた顔方向の認識システムの構築

抄録

This paper proposes a new method of recognizing a facial orientation based on tracking a pair of nostrils. Due to the fact that nostril shape is stable compared with other facial parts such as eyes or mouth, the nostril image was treated as a feature point. USB camera was located in front of the user’s face and the obtained face image was taken in real-time for image processing. The face orientation could be detected according to the threshold circumstances. The constructed system was applied to the selecting operation for auto-wheelchair running. Here, we consider the setting of the starting position and goal position by turning to the left or right directions. Nodding action was used for decision of starting position and goal position. As a result, the user could operate only by face direction without handling.