电子测量与仪器学报

尹柏强,邓影,王署东,胡增超,李兵,佐磊.时频广义 S 变换和 VL-MOBP 神经网络在人体动作识别中的应用[J].电子测量与仪器学报,2020,34(11):1-9

时频广义 S 变换和 VL-MOBP 神经网络在人体动作识别中的应用

Application of time-frequency generalized S transform and VL-MOBP neural network in human motion recognition

DOI：

中文关键词: 时频广义 S 变换 VL-MOBP 神经网络表面肌电信号动作识别

英文关键词:time-frequency generalized S transform VL-MOBP neural network surface electromyographic action recognition

基金项目:国家自然科学基金(61971175)、国家重点研发计划“重大科学仪器设备开发”项目(2016YFF0102200)、国家自然科学基金重点项目(51637004)、中央高校基本科研业务费(JZ2019YYPY0025)资助项目

作者	单位
尹柏强	1.合肥工业大学电气与自动化工程学院
邓影	1.合肥工业大学电气与自动化工程学院
王署东	1.合肥工业大学电气与自动化工程学院
胡增超	1.合肥工业大学电气与自动化工程学院
李兵	1.合肥工业大学电气与自动化工程学院
佐磊	1.合肥工业大学电气与自动化工程学院

Author	Institution
Yin Baiqiang	1.School of Electrical and Automation Engineering, Hefei University of Technology
Deng Ying	1.School of Electrical and Automation Engineering, Hefei University of Technology
Wang Shudong	1.School of Electrical and Automation Engineering, Hefei University of Technology
Hu Zengchao	1.School of Electrical and Automation Engineering, Hefei University of Technology
Li Bing	1.School of Electrical and Automation Engineering, Hefei University of Technology
Zuo Lei	1.School of Electrical and Automation Engineering, Hefei University of Technology

摘要点击次数: 215

全文下载次数: 374

中文摘要:

针对仿生假肢动作识别问题,提出基于时频广义 S 变换和 VL-MOBP 神经网络的下肢动作识别方法。首先用时频广义 S 变换对年龄在 20~ 40 岁,身高在 170~ 185 cm,体重在 50~ 75 kg 的 22 名男性测试者下肢 4 种表面肌电信号和膝盖弯曲度信号进行多分辨率分析,得到在时间和频率分辨率较好情况下信号时频累计特性曲线,然后提取时频累计特性曲线幅值的均值和标准差作为特征向量,用 VL-MOBP 神经网络对人体下肢的行走、站立及静坐 3 种动作进行识别。实验结果表明,提出的下肢动作识别方法能够取得很好的识别效果,平均识别准确度达 96. 67%,高出小波变换约 56%,高出短时傅里叶变换约 36%,验证了该方法在动作识别中的有效性。

英文摘要:

Aiming at the needs of bionic prosthetic motion recognition, a lower limb motion recognition method based on time-frequency generalized S transform and VL-MOBP neural network was proposed. First, time-frequency generalized S-transform was used to measure 4 kinds of surface electromyographic signals and knee flexion of the lower extremities of 22 male subjects aged between 20 and 40 years old, between 170 cm and 185 cm tall and weight between 50 kg and 75 kg. Using multi-resolution analysis of the frequency signal to obtain the time-frequency cumulative characteristic curve of the signal when the time and frequency resolution were good, then extracting the mean and standard deviation of the amplitude of the time-frequency cumulative characteristic curve as the feature vector, and using the VL-MOBP neural network to recognize the three movements of human lower limbs: Walking, standing, and sitting. The experimental results showed that the proposed lower limb movement recognition method can achieve good recognition results, with an average recognition accuracy of 96. 67%, which is about 56% higher than the wavelet transform and about 36% higher than the short-time Fourier transform. Effectiveness in motion recognition has been verified.

查看全文查看/发表评论下载PDF阅读器