期刊
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
卷 71, 期 -, 页码 -出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIM.2022.3217868
关键词
Armband; hand gesture recognition; high-density surface electromyogram (sEMG); measurement location; prosthesis
资金
- National Natural Science Foundation of China [62173094]
- Shanghai Municipal Science and Technology International Research and Development Collaboration Project [20510710500]
- Natural Science Foundation of Shanghai [20ZR1403400]
This study investigates the application of sEMG-based hand gesture recognition for prosthesis or armband in the human-machine interface. The experiment finds that the measurement location greatly influences the accuracy of gesture recognition. Therefore, the study provides a factor-screening tool to help users customize their systems according to their physical conditions and requirements.
Surface electromyogram (sEMG)-based hand gesture recognition for prosthesis or armband is an important application of the human-machine interface (HMI). However, the measurement location of sensors greatly influences the hand gesture performance, especially with the interday or intersubject validation protocols. Therefore, we acquired two-day hand gesture data of 41 subjects with a 256 (16 x 16) channel high-density sEMG electrode array. With the acquired data, we initially compared the support vector machine (SVM) and other four state-of-art classifiers under three validation protocols, i.e., intraday, interday, and intersubject validation protocols. Then, we screened 14 feature optimization techniques, including five feature-projection methods and nine feature-ranking approaches. To present the accuracy tendency with varying measure locations, we systematically explored the ten-hand gesture performance using data of 16 prosthesis measurement locations (PMLs) and 15 armband measurement locations (AMLs). As a result, the SVM classifier was suitable for the intraday and interday validation protocols and the 2-D convolutional neural network was selected for the intersubject validation protocol. The mean accuracies of the hand gesture classification ranged from 95.68% to 99.12% (intraday validation), from 68.41% to 88.02% (interday validation), and from 63.39% to 86.33% (intersubject validation) for the prosthesis application. In addition, for the armband application, the mean accuracies ranged from 96.25% to 97.43% (intraday validation), from 67.44% to 75.83% (interday validation), and from 65.53% to 75.40% (intersubject validation). The accuracy is greatly correlated with the measurement location, which is highly associated with the neuromuscular structures of human bodies. In summary, our work can serve as a factor-screening tool for users customizing their systems according to their physical conditions and requirements.
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