SIA OpenIR  > 机器人学研究室
Alternative TitleThe Method for Gestures Recognition Based on Myo Rotation Shifts Estimation and Adaptive Correction
李自由1,2,3; 王丰焱1,2,3; 赵新刚1,2; 丁其川4; 张道辉1,2; 韩建达1,2,5
Source Publication自动化学报
Indexed ByEI ; CSCD
EI Accession number20204509468638
Contribution Rank1
Funding Organization国家自然科学基金(61573340, 61773369, U1813214)
Keyword表面肌电信号 电极偏移 自适应校正 模式识别

在基于表面肌电信号的手势识别系统中,针对Myo环形电极多次实验间旋转位置不同导致的识别精度降低问题,提出了种基于极坐标系的电极位置偏移估计与自适应校正的识别方法。该方法首先建立相对于环形肌电传感器的极坐标系,提出了极坐标系下活跃极角(Activation Polar Angle, APA),用于估计实验中传感器相对于初始位置的横向旋转偏移角度,迸而建立基于偏移角度的线性变换模型,在肌电信号特征空间内,对电极偏移位置下的样本进行自适应校正。在8 种常用手势识别应用中,设计了两种实验范式利用传感器各通道数据循环平移模拟电极横向旋转偏移实验和肌电传感器在小臂肌肉上的真实旋转偏移实验。结果均表明所提出方法的识别精度远高于未进行校正的模型识别精度。因此,所提出的电极偏移估计与自适应校正识别方法,不仅有效提高了表面肌电交互系统识别的鲁棒性,也降低了使用者在多次使用时训练成本与学习负担。

Other Abstract

In the gestures recognition system based on sEMG, an electrode shifts estimation and adaptive correction solution is proposed, to the problem of low recognition accuracies interfered by the Myo armband ring-electrode rotation shifts among experiments. The proposed method first estiblishes a polar coordinate system that is stationary relative to the ring sEMG acquisition system, and defines an activation polar angle (APA) that will be later used for measuring the electrode rotation shifts relative to its initial position. Based on the measurement of electrode shifts, interfered sEMG samples are adaptively corrected by a linear transformation in feature space. In the sEMG-based recognition application of eight common gestures, two experimental paradigms are conducted, including a simulation rotation paradigm by right shifts of eight-channel sEMG signals and a real one by shifts of the ring acquisition system on the lower arm. Results of both experimental paradigms dimenstrate that the recognition accuracies of our proposed method are much higher than those of non-corrected models. In conclusion, the proposed electrode shifts estimation and adaptive correction method, improves the robustness of sEMG-based recognition system, and reduces the training time and learning burden for users.

Citation statistics
Document Type期刊论文
Corresponding Author赵新刚
Recommended Citation
GB/T 7714
李自由,王丰焱,赵新刚,等. 基于Myo旋转偏移估计与自适应校正的手势识别方法[J]. 自动化学报,2020,46(9):1896-1907.
APA 李自由,王丰焱,赵新刚,丁其川,张道辉,&韩建达.(2020).基于Myo旋转偏移估计与自适应校正的手势识别方法.自动化学报,46(9),1896-1907.
MLA 李自由,et al."基于Myo旋转偏移估计与自适应校正的手势识别方法".自动化学报 46.9(2020):1896-1907.
Files in This Item:
File Name/Size DocType Version Access License
基于Myo旋转偏移估计与自适应校正的手势(1708KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[李自由]'s Articles
[王丰焱]'s Articles
[赵新刚]'s Articles
Baidu academic
Similar articles in Baidu academic
[李自由]'s Articles
[王丰焱]'s Articles
[赵新刚]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[李自由]'s Articles
[王丰焱]'s Articles
[赵新刚]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 基于Myo旋转偏移估计与自适应校正的手势识别方法.pdf
Format: Adobe PDF
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.