即時物件切割技術之研究

• Main Authors: Sheng-Yi Chiu, 邱勝鎰
• Other Authors: Chung-Cheng Chiu
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/30537831435499080323

碩士 === 國防大學理工學院 === 電子工程碩士班 === 98 === Traditional surveillance system takes a lot of manpower, and its reliability of human decision may be inaccurate easily when human are wearied. These problems help the intelligent surveillance system to develop rapidly. And the advancement of computer hardware also speeds up the development of intelligent surveillance systems. The most important part for the intelligent image surveillance system is to segment the moving objects immediately, and to use the segmented object applied in real-world systems, such as the communications and transportation system or the secure intrusion detection system and so on. The algorithm in this thesis can segment the moving objects immediately and get the initial background quickly. According to our experiments, the proposed algorithm can obtain better segmented objects under various environments. In the thesis, we propose an algorithm applied in video sequences through a static camera. The algorithm can classify and extract the background effectively. Even in slow moving objects in the image sequence, the proposed algorithm can get background accurately. Then we use the background subtraction method to segment the objects from input images. And we solve the problem of object under the interferences of auto white balance of cameras.