Observing passing people by using fiber grating vision sensor

Information obtained by using the monaural camera is 2D data. Therefore, it is difficult to get information of the person's position, direction, and shape. Then, the sensor, which is able to obtain 3D data, is required. In this paper, we propose a method of observing the position, direction and shape of people by using 3D information. In this research, people's 3D information is acquired using FG vision sensor, which has been developed by the authors. Position information, direction information and height information are detected by extracted a cylinder from 3D data. In this process, Hough transform is used to detect position, X-axis and Y-axis of ellipse of a cylinder. The tracking is carried out using the position and the height information in real-time.