A Fast Method for Building and Updating Background Model

Abstract

On the basic of the video traffic surveillance system, the Vehicle detection is a crucial step. A typical method is background subtraction. Extracting and updating the background plays an important role on speed and efficiency of detection. For this reason, the author puts forward a fast and effective method for building and updating the background model. First, adopt the improved statistical method to build initial background. Second, get the vista of the picture per frame by means of background subtraction, and bring forward the background updating method based on the container technology. Finally, in its updating stage, use the pixel excluding the pixel in vista region to renew the background model. The data of experimental picture states clearly that competed with the traditional method, detection technology is robust and accurate. Furthermore, on the realization of arithmetic, it is easier to operate and it has higher timeliness.