基于多特征融合的健美操手臂动作轨迹识别  

Recognition of Aerobics Arm Movement Trajectory Based on Multi Feature Fusion

作　　者：李严[1] LI Yan(City College of Xi'an Jiaotong University,Xi'an 710018 China)

机构地区：[1]西安交通大学城市学院,陕西西安710018

出　　处：《自动化技术与应用》2023年第9期90-93,共4页Techniques of Automation and Applications

摘　　要：由于传统的健美操手臂动作轨迹识别方法存在识别结果不准确等问题,提出一种基于多特征融合的健美操手臂动作轨迹识别方法。在深度图像序列中获取健美操运动员的手臂关节点信息,通过关节点坐标提取位移特征和部件中心特征。利用空间区域分块方法进行动作图像的手臂动作空间规划,构建手臂区域定姿模型。通过模板自匹配和小波多尺度分解方法对提取到的特征进行轨迹边缘检测,实现手臂动作轨迹识别。仿真实验结果表明,所提方法能够获取高精度的健美操手臂动作轨迹识别结果,有效减少漏识率和识别费用。Because the traditional aerobics arm motion trajectory recognition method has problem such as inaccurate recognition results,a aerobics arm motion trajectory recognition method based on multi feature fusion is proposed.The arm joint point information of Aerobics athletes is obtained in the depth image sequence,and the displacement feature and component center feature are extracted through the joint point coordinates.The arm motion space planning of motion image is carried out by using the spatial region segmentation method,and the arm region pose determination model is constructed.Through template self matching and wavelet multi-scale decomposition,the trajectory edge of the extracted features is detected to realize arm motion trajectory recognition.Simulation results show that the proposed method can obtain high-precision aerobics arm motion trajectory recognition results,and effectively reduce the missing recognition rate and recognition cost.

关 键 词：多特征融合 动作轨迹识别 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]