Research On Fast Tracking Algorithm For Moving Object Based On Optical Flow

Target tracking has a broad application prospect in military guidance, securitymonitoring, medical diagnostics, and many other fields. The main purpose of targettracking is to obtain the motion and trajectory parameters of the target, the difficulty oftarget tracking lies in the target and the scene where the target lies are always changingduring the process of motion, in addition, the target obscured and other issues will affectthe tracking results. The main purpose of this paper is to achieve the target tracking byusing optical flow method, and apply it to the laboratory teaching experiment platform.First of all, this paper presents a brief tracking method about space rigid body,when we constrain the rigid body do fixed-axis rotation, we can get the motionparameters of the target in the two-dimensional plane by matching four points, the speedof the tracking algorithm will make a great improved. Secondly, when we use opticalflow method for point tracking, the problem is that the optical flow method are greatlyinfluenced by the environmental changes, such as when the target obscured or thesurface of target illumination changes, which will cause the failure of tracking. Aimingat the problem of optical flow method are susceptible to interference, this paper presentsa filtering method based on probability, the motion parameters that larger or manifesterror deviation values can be filtered, and finally calculate a reliable parameter values.Finally, in this article, we verify the effectiveness of the tracking method throughexperiment and compare it with other common tracking method to verify the methodused in this article on the advantage of quickness.