排球运动员手臂运动轨迹优化识别仿真研究

摘要

对于排球运动员手臂运动轨迹的优化研究可以有效提高运动员在比赛中扣球质量.进行挥臂动作轨迹的优化时需要进行动态手臂跟踪,将具有三维空间特征的运动轨迹转化为一维的运动轨迹.传统方法则忽略了该转化过程,直接对三维空间轨迹进行分析,导致轨迹优化过程复杂且优化效果不好.提出基于混沌理论的排球运动员手臂运动轨迹优化识别方法.以背景差分原理为依据检测运动员运动轨迹,利用颜色直方图的粒子滤波进行动态手臂跟踪,融合于混沌理论进行运动员手臂运动轨迹的相空间重构,从重构的相空间提取代表运动员手臂运动轨迹的混沌不变量,并将具有三维空间特征的手臂运动轨迹转化为一维的手臂运动轨迹,并完成了对排球运动员手臂运动轨迹优化识别.仿真结果表明,所提方法识别精确度高,为提升排球运动员扣球技术提供了有力的科学依据.%In this paper,we proposed an optimization recognition method for arm movement trajectory of volleyball player based on the chaos theory.Firstly,we detected the athlete movement trajectory according to the background subtraction theory and used the particle filter of color histogram to make dynamic arm tracking.Then,we made the phase-space reconstruction to arm movement trajectory integrated with the chaos theory and extracted chaos invariant representing movement trajectory of athlete arm from the reconstructed phase-space.Moreover,we converted the arm movement trajectory having 3D space feature into one-dimensional movement trajectory.Finally,the optimization recognition on the arm movement trajectory of volleyball player was completed.The simulation results show that the method has higher recognition precision,and it can provide a powerful scientific basis for improving smash technology of volleyball player.