The tracking technology of the 3D space position of the target is of great use to the analysis and understanding of professional volleyball match, but there are many factors that affect the volleyball tracking rate in the competition. In this paper, the image processing technology is combined with 3D spatial matching technology, and a sports volleyball tracking system framework is established with particle filter, which improves the accuracy of volleyball tracking more effectively. This article begins with a 3D tracking volleyball workflow diagram and selects using particle filters to improve tracking success rates. Secondly, the paper puts forward the similarity estimation method according to the characteristics of volleyball itself, mainly the similarity estimation of the base and HSV color space, to solve the problem of the super-computing of real-time tracking. Finally, the results of the simulation experiment can be seen that the method can greatly improve the tracking success rate, high efficiency and high accuracy of the characteristics of the dynamic video sports volleyball tracking, has an unpredictable application development value.
© Copyright 2020 IEEE Access. Alle Rechte vorbehalten.