Research Of Motving Object Detection Algorithm In Video Surveillance

With the rapid development of science and technology, Intelligent monitoring systems are more widely used. Intelligent monitoring system is made of moving object detection, object tracking, feature extraction and be havior analysis and recognition. Moving object detection is not only the premier Intelligent monitoring system processing, but also the most critical par t.The quality of object detection has an important influence on subsequent processes. Thus, real-time and high accuracy rate moving object detection algorithm has been a hot research topic in scientists.On the basis of deep research on video image preprocessing and some of the classic algorithm for moving object detection, this paper analyses their characteristics and advantages and disadvantages of various classic algorithms such as optical flow method, the inter-frame difference method and background difference method etc. The inter-frame difference method and background difference method is the most widely used moving object detection algorithm, the inter-frame difference method has a good adaptability to the dynamic background environment, but the detection result is easy to appear empty area and object Edge is always not complete. Background difference method can better detect body contour line, but dynamic environmental factors such as light, the weather changes can have a big impact on it. Therefore this paper designs a new method Combining inter-frame difference method and background difference method to detect moving object. The purpose of this new method is to ensure the integrity of moving object information and reduce the influence of noise from dynamic environment at the same time.The shadow of moving object have a great influence to the object shape,so the shadow must be removed. This paper analyzes the principle of removing shadow in HSV color space. According to the method of its implementation, this paper carry on the Simulation experiment. In HSV color space, The brightness value of shadow area has changed greatly in contrast with the brightness value of background area. Therefore, comparing the brightness value can determine which pixel is located in the shadow area. The simulation experiment results show that shadow remov ed method in HSV space can achieve a better shadow suppression effect.