Real-Time Motion Detection and Automatic Motion Tracking System

• Main Authors: Chien, Lung-Chih, 簡隆至
• Other Authors: Tsai, Chau-Ren
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/27256664155220012685

碩士 === 國立臺灣科技大學 === 電機工程系 === 92 === The computer vision related research already carried on had in more than 30 years. Motion detection and motion tracking was always one of computer vision related research important projects, and related technology also in abundance was proposed, and has obtained the suitable degree the result. However the robust motion object detection and motion tracking still is an extremely difficult task. In the present paper, we will use Moving Edge Detection and Mean Shift algorithm to build motion detection and motion automatic tracking system based on the computer vision. In present paper, after we pick up from digital camera to the run-on time images, we then can detect the motion object sub-area the position by Moving Edge Detection, then uses this motion object sub-area the position to establish model image of the motion object. After model image of the motion object establishment, we then can use the Mean Shift algorithm to track the model image of the motion object, and penetrates the RS-232 in series communication to control the active digital camera to carry on the rotation, causes the motion object sub-area to be allowed continually to maintain at the picture central position, then achieves motion object tracking the task.