Design Of Target Detection And Tracking System Based On Improved Optical Flow Field

The detection and tracking based on moving objects is a common problem in road traffic regulation.We usually use pattern recognition to track vehicles in video images.Vehicle and other target tracking is in each image of the video image to determine the target location,so we need to achieve the vehicle tracking.With the continuous development of science and technology,target detection and tracking technology more and more intelligent.The detection and tracking algorithm based on moving target has wide application prospect in many fields such as virtual reality,process control,military,medical,road traffic monitoring and so on.The key to the tracking algorithm is how to quickly and efficiently achieve automatic target positioning in each frame of the image.The algorithm based on optical flow is widely used in target detection and tracking system.In this paper,several commonly used target detection and target tracking algorithms are introduced.By comparing the advantages and disadvantages between the algorithms,an improved target detection and tracking algorithm is proposed.Finally,the simulation experiment is carried out.The algorithm has a certain accuracy.In this paper,the main research contents and innovations of this paper are discussed.Three classical moving target detection algorithms,BDM(background difference method),FDM(frame difference method),OF(optical flow method)[4]and the improvement of moving target detection and tracking based on optical flow method.The basic principle of the three algorithms is briefly introduced,and the advantages and disadvantages of each algorithm are compared.BDM detection is very sensitive to the detection of its dynamic background changes;FDM although its calculation is smaller than other algorithms,but when used to detect the moving target will occur when the "hole" phenomenon;OF in complex environmental conditions can be carried out motion target detection,but easy to receive noise interference.At the same time,the generalized optical flow algorithm is time-consuming and can not meet the real-time requirement,and can not accurately detect the shape of the moving target.In this paper,we propose a new type of optical flow calculation model based on COF(Characteristic optical flow method)and ViBe algorithm to accurately detect the moving objects in the video image to obtain the dense texture flow of the image.Edge information and optical flow of information for multi-feature fusion,from the calculation of the accuracy and robustness of the two aspects of the assessment.For the tracking of moving objects,this paper mainly adopts the tracking method of EKF filter based on optical flow.The EKF filter algorithm can solve the problem of tracking the target under nonlinear observation.When the moving target encounters occlusion,the EKF filter can estimate and predict the moving target well.Through the analysis of Matlab simulation results,EKF filter technology to track the moving target,its operation range is broader,faster calculation results more accurate.