Intelligent Video Surveillance System Design

Video surveillance has been widely used in many fields, such as national defense, the traffic control, the intelligent public security and so on. Video surveillance system nowadays, however, depends on manual operation, which wastes resources and influences the efficiency. Concerning about the improvements of efficiency and robustness of the system, in this thesis, an intelligent video surveillance system is designed and implemented, in which the key techniques such as automatic diaphragm and autofocus, moving object detection and tracking are integrated.MCS (Mounting Climbing Servo) method based on image differencing is discussed and used in autofocus, and a combined method of Sobel and SMD is implemented. On the research of moving object detection, temporal differencing algorithm and background subtraction algorithm are studied. By many experiments, the detection results using different methods are analyzed and compared under different environments. Following, object tracking using mean shift is studied. At last, the software, by which we can detect and track moving objects efficiently and effectively, has been developed.