Research On Tracking And Positioning Technology Of Moving Targets In Vedio

Moving targets tracking has been an important research topic in the area of computer vision , digital video , image processing and pattern recognition .The key that lies in the realization of tracking is right to select target characteristics, reasonable to select feature matching algorithm and accurate to estimate the state of the target . In this paper , we conduct an in-depth study of the video moving target tracking algorithm , with the car moving on the road as a research object .First, we introduce the main application of target tracking , the research and development status of tracking algorithm , and the basic tenets of target tracking .We then discuss the principles of relevant tracking and conduct a simulation experiments . The correlation tracking algorithm based on the function of the minimum absolute difference is analyzed , the comparison of two algorithm's performance in target tracking is presented . To hurdle the limitations of the fixed template , we study a cross-correlation tracking algorithm based on variable templates . The experimental results show that this method is superior to the correlation tracking algorithm based on the function of the minimum absolute difference .We construct a tracking framework based on particle filter and combine it with relevant tracking algorithm , educing a correlation tracking algorithm based on particle filter . Experimental results show that the algorithm can effectively overcome the shielding problems of barriers and improve the robustness of the target tracking .Finally , we discuss the basic principles of Kalman filtering and the Kalman filter tracking model , and construct a tracking algorithm based on the Kalman filter tracking model . We extend the moving objects from a single car to two moving cars . Simulation results show that the algorithm can track the two moving targets effectively .