首页|基于改进MEMS惯性传感器的人体动作智能捕捉

基于改进MEMS惯性传感器的人体动作智能捕捉

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
  • 维普
由于MEMS惯性传感器在人体动作数据采集时会受到重力加速度的影响，导致人体动作捕捉效果差，提出基于改进MEMS惯性传感器的人体动作智能捕捉方法。运用MEMS惯性传感器获取人体动作数据，并对采集的数据实施数据去噪、误差校正、重力加速度去除等处理，优化MEMS惯性传感器的数据采集效果，获取实际的人体运动数据。依据获取的人体运动数据，将方差作为人体整体运动幅度的特征向量值，提取人体动作特征，结合支持向量机对动作进行分类识别，实现人体动作捕捉。实验结果表明，使用上述方法开展人体动作捕捉时，捕捉精准度较高、捕捉时间较短，且实际效果较好。
Intelligent Human Motion Capture Based on Improved MEMS Inertial Sensor
Due to the influence of gravity acceleration on the collection of human motion data using MEMS iner-tial sensors,the effectiveness of human motion capture is poor.Therefore,an intelligent human motion capture method based on improved MEMS inertial sensors was proposed.Firstly,MEMS inertial sensor was used to obtain human mo-tion data.For the collected data,data denoising,error correction and gravity acceleration removal were carried out.And then,the data collection effect of MEMS inertial sensor was optimized to obtain actual human motion data.Ac-cording to the data,the variance was taken as the feature vector value of the motion amplitude of human body.Final-ly,the features human action was extracted.Meanwhile,the action was classified and recognized with the support vec-tor machine.Thus,the human motion capture was achieved.Experimental results show that the capture accuracy is higher,and the capture time is shorter.In addition,the actual effect is better during the human motion capture.

Inertial sensorHuman motion captureHuman motion featureSupport vector machineError correc-tionGravitational acceleration

杨艺、贺晓阳

展开 >

大连工业大学,辽宁 大连 116034

大连工业大学信息科学与工程学院,辽宁 大连 116034

惯性传感器 人体动作捕捉 人体动作特征 支持向量机 误差校正 重力加速度

教育部人文社会科学研究项目(综合研究类)教育部科技发展中心高校产学研创新基金"北创助教"基金立项辽宁省教育厅高校基本科研项目

21YJCZH2162018C01005LJKFR20220223

2024

计算机仿真
中国航天科工集团公司第十七研究所

计算机仿真

CSTPCD
影响因子：0.518
ISSN：1006-9348
年,卷(期)：2024.41(1)
  • 16