Surveillance Video Of Moving Objects

Intelligent video surveillance system is a new kind of digital surveillance system, which based on the analysis and operation of video content. This paper mainly worked on the basic part of intelligent digital video surveillance system: moving objects detection and tracking.Background subtraction is a commonly used moving objects detection method, it subtracts the background model which is stored beforehand or observed in time from the current frame, by threshold operation getting moving objects. The abstraction and update of background, the effect of background disturb, illumination changes and shadows are the problems that must be considered in background subtraction method. This paper presents a novel moving objects detection method, which addresses all the problems presented above effectively. A statistical method is used to obtain the background model, which is updated real time in order to adapt to illumination changes and scene changes. After threshold operation, morphologic operation and connected region area measurement are introduced to solve background disturb problem. At last, shadows are detected using HSV color space information and gets exact moving regions.Tracking is based on moving objects detection, its main purpose is to get the tracks and motion feature such as velocity and direction of the moving object. An effective moving objects tracking method should be able to track multiple objects at one time, and should be able to deal with objects overlapping, merging and splitting. This paper present a robust moving objects tracking method, we consider the moving objects with their outside rectangular, track them in the scene, and get their tracks and moving parameters.At last, we count the number of moving objects that pass through the surveillance region on the base of detection and tracking.To validate the correctness of our detection and tracking method, the author builds a prototype system, and tests it in different environment. Experiment results show that presented method is effective.