期刊
TRAITEMENT DU SIGNAL
卷 38, 期 3, 页码 689-697出版社
INT INFORMATION & ENGINEERING TECHNOLOGY ASSOC
DOI: 10.18280/ts.380316
关键词
wavelet thresholding; limb rehabilitation; electromyography (EMG) signal; pattern recognition
资金
- Department of Education [JJKH20200569KJ]
This study explores the identification and analysis of limb rehabilitation signal based on wavelet transform, proposing a limb EMG pattern recognition method and utilizing support vector machine to evaluate the rehabilitation status of patients. Experimental results demonstrate the effectiveness of the proposed method in noise removal and EMG signal recognition.
The development of science and technology has promoted the extensive application of surface electromyography (sEMG) collection technique in real-time exercise testing, assistive judgment of rehabilitation therapy, and assessment of intelligent artificial limb application. However, there is a severe lacking of studies on pattern recognition based on effective signal, and evaluation of limb rehabilitation status. To make up for the gap, this paper explores the identification and analysis of limb rehabilitation signal based on wavelet transform. Specifically, the authors detailed the basic flow of sEMG signal generation in motor unit during limb rehabilitation exercise, and proposed a limb EMG pattern recognition method. Then, support vector machine (SVM) was selected to recognize the pattern of the EMG signal extracted from the limb rehabilitation exercise of patients, and to judge the rehabilitation status. Finally, wavelet thresholding was combined with total variation denoising (TVD) to effectively remove the noise from EMG signal. The proposed method was proved effective through experiments.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据