Research On Moving Objects Detection In Video-based Surveillance System

In the scene under surveillance,the basic issue of video frequency surveillance technic is the research of the motion detection under a static background.It is on the bottom of the video frequency systemic and is the basic of subsequent advanced application.The research of the motion detection under a static background is very important for the systemic of the video frequency surveillance.A familiar instance and also the precondition of my task is the instance when the pickup camera is still and the foci of the camera lens is changeless.In this instance,there are often three method to check:optical flow,temporal difference,background subtraction.Based on studying and organizing the relative information,this discourse includes the detecting the moving objects under a static background.The method of optical flow is comparatively complex and it's real time is not very good.The method of temporal difference has a very good adaptability to dynamic circumstance.But,it has a request to the speed of the moving objection when we use it,so this method is restricted when we use it.Temporal difference as a efficacious method,is the primary content of the discourse.We will detailedly research the method of time average and pick-up background base on gauss model(s).I will use two chapters to detailedly research the method of background reconstruction algorithm based on pixel intensity classification(PCI) and pixel Intensity Classification and Mend(PCIM).According to the hypothesis that the background pixel intensity appears in image sequence with maximum probability,the pixel intensity differences between sequential two frames are calculated,and the intensity values at the pixels are classified by means of these differences.For the new algorithm,neither the pre-training of the background without any moving target,nor the models of background and targets are needed.Simulation results indicate that background can be reconstructed correctly by using the new algorithm,so the target can be extracted perfectly and tracked successfully.In the last chapter,we will compare and analyse these methods.