| 张 娜,张明进,王晓冬,梁 铁,李 俊,熊 鹏,刘晓光.基于表面肌电信号的手指关节角度估计方法[J].电子测量与仪器学报,2023,37(8):60-70 |
| 基于表面肌电信号的手指关节角度估计方法 |
| Estimation of finger joint angles based on surface electromyographic signal |
| |
| DOI： |
| 中文关键词: 表面肌电信号 深度森林 人工神经网络 通道注意力机制 手指关节角度估计 |
| 英文关键词:sEMG deep forest artificial neural network channel attention mechanism finger joint angle estimation |
| 基金项目:国家自然科学基金(62276087)、河北省自然科学基金(2021201002)、河北省教育厅科技计划项目(ZD2020146)资助 |
|
|
| Author | Institution |
| Zhang Na | 1. China Key Laboratory of Digital Medical Engineering, Hebei University,2. Stone pharmaceutical Group Zhongnuo Pharmaceutical Co. , LTD. |
| Zhang Mingjin | 1. China Key Laboratory of Digital Medical Engineering, Hebei University，3. College of Electronic Information Engineering, Hebei University |
| Wang Xiaodong | 4. Affiliated Hospital of Hebei University |
| Liang Tie | 1. China Key Laboratory of Digital Medical Engineering, Hebei University，3. College of Electronic Information Engineering, Hebei University |
| Li Jun | 1. China Key Laboratory of Digital Medical Engineering, Hebei University，3. College of Electronic Information Engineering, Hebei University |
| Xiong Peng | 1. China Key Laboratory of Digital Medical Engineering, Hebei University，3. College of Electronic Information Engineering, Hebei University |
| Liu Xiaoguang | 1. China Key Laboratory of Digital Medical Engineering, Hebei University，3. College of Electronic Information Engineering, Hebei University |
|
| 摘要点击次数: 494 |
| 全文下载次数: 450 |
| 中文摘要: |
| 为了实现智能假手能够自然地模拟人手的连续运动,提出了基于 sEMG 的 DF-ANN 模型来估计手指关节角度的方法。
该方法引入了通道注意力机制中的 SE-Net 模块增强了 sEMG 的相关特征表达,减少 sEMG 重要特征的损失,有效提高了回归模
型的性能,选取 10 名健康的受试者进行 10 种不同手势的实验,选择 R-Squared(R
2
)等回归衡量指标来评估该方法关节角度估
计的精度,实验结果显示 R
2 为 86. 5%。 与未引入 SE-Net 的 DF-ANN 模型,单独的深度森林和人工神经网络相比,R
2 大约提高
了 4%。 这表明该方法能够有效减小 sEMG 的关节角度连续解码的误差,能够有助于实现智能假手的柔顺控制。 |
| 英文摘要: |
| In order to achieve an intelligent prosthetic hand that can naturally simulate the continuous motion of a human hand, this
paper proposes a DF-ANN model based on sEMG to estimate the finger joint angle. The method introduces the SE-Net module in the
channel attention mechanism to enhance the relevant feature expression of sEMG, reduce the loss of essential features of sEMG, and
effectively improve the performance of the regression model. 10 healthy subjects were selected for experiments with 10 different hand
gestures, and regression measures such as R-Squared (R
2
) were chosen to evaluate the accuracy of the method’s joint angle estimation.
The experimental results showed an R
2
of 86. 5%. Compared with the DF-ANN model without introducing SE-Net, the deep forest, and
an artificial neural network alone, the R
2
is improved by about 4%. It indicates that the method effectively reduces the error of successive
decoding of joint angles of sEMG and can contribute to the supple control of intelligent prosthetic hands. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
