基于形体轮廓特征提取的运动动作标志分析

Movement Action Mark Analysis Based on Body Contour Feature Extraction

对健美操等运动状态下的标志性动作分析时评价和刻画健美操运动训练和比赛水平的定量性标准。传统的运动动作标志性分析方法采用运动幅度检测方法,对健美操这类运动幅度呈现不规则差异的运动项目来说并不适用。基于图像处理的方法,提出一种基于形体边缘轮廓特征点提取的运动动作标志性分析算法,进行了形体边缘轮廓特征点数据采集系统设计。提取形体边缘轮廓特征点,将空间邻域信息融入到运动形体图像的标志性动作检测中,采用亮点模型衍射的方法实现视觉穿透,搜寻健美操运动员的技术特征,实现对健美操运动动作的标志性分析算法改进。仿真实验结果得出,采用该算法能有效准确地提取健美操的形体边缘轮廓特征点,运动动作标志性检测概率较高,性能稳健,具有优越性。研究成果在健美操形体评价等领域中具有较好的应用价值。

Aerobics symbolic action analysis under motion condition assessment and characterization of standard quantitative aerobics training and competition level. Analysis of applying the method of range of motion mark movement tradition, it is not suitable for sports aerobics this kind of range of motion presented irregular difference. Based on image processing method, put forward a kind of analysis algorithms for sign of movement body edge contour extracting feature points based on the designed shape edge features of data acquisition system. The extraction of feature edge contour feature points, the spatial neighborhood information into the movement of body image landmark action detection, using the method of highlight model diffraction realize visual penetration, search technology characteristics of aerobics athlete, can achieve the improvement of analysis algorithm of symbol of aerobics movement of. The simulation experiment results, the algorithm can effectively and accurately extract the feature edge contour feature points of the aerobics movement marked a high detection probability, stable performance, has the superiority. The research results have good application value in the fields of Aerobics physique evaluation.