Moving Target Tracking System Based On Ultrasonic Motors

Detection and tracking of moving targets are key to the digital video analysis, and are the hotspotof the certain research domain, which have around increasing attention at home and aboard.It haswide applications in smart surveillance systems, next generation computer interfaces and many otherfields in need of such automation. The objective of this paper is to optimize the selection of the initialtarget and increase the precision of the tracking contrail by the improved algorithm of object tracking，which is called the algorithm of the particle filter based on Mean-Shift. It is capable of keeping thetargets being tracked in the center of the frame. The research achievements mainly are as followings:Firstly, the status quo and research methods of object tracking were analyzed and summarized.And many common algorithms of target detection are introduced. The method of mixed-Gaussbackground model based on color-background is adopted.Secondly, the algorithm of Mean-Shift and the algorithm of particle filter are introduced. At thesame time, the particle filter algorithm would be combined with Mean-shift algorithm. The diversityis maintained and the time is reduced due to litter particle.Thirdly, the pan-tilt can rotate in horizontal and vertical directions by using double ultrasonicmotors served as the driving units. The controlling of the ultrasonic motors achieves the desired resultby the fuzzy control algorithm.Finally, the experiments based on target tracking were done. A series of experiment results anddata analysis find the method’s validity and practicability. The program developed with Visual C#isable to track the moving target in real-time and multi-objective existence.