摘要

The purpose of this paper is to apply human behavior recognition technology based on support vector machine theory in Wushu training. This paper presents descriptions of the actions being performed by a human in a series of videos. A Weizmann database is adopted to achieve the output, which comprises videos with a homogeneous and static background. Each video contains a single periodic action that is performed by a single person. Experiment results demonstrate that the human behavior recognition technology based on support vector machine theory can improve the efficiency and performance of correct actions in the Wushu Training process.