Abstract
The surface electromyogram (sEMG) contains a wealth of motion information, which can reflect user's muscle motion intentions. The decoding based on sEMG has been widely used to provide a safe and effective human-computer interaction (HCI) method for neural prosthesis and exoskeleton robot control. The motor intention decoding based on low sampling frequency sEMG may promote the application of wearable low-cost EMG sensors in HCI. Therefore, a motor intention decoding scheme suitable for low frequency EMG signal is proposed in this paper, that is, transfer learning based on Alexnet. Moreover, the effects of different feature extraction methods and data augmentation with Gaussian white noise are fully analyzed. The proposed algorithm is evaluated with the NinaPro database 5. The highest accuracy can reach 70.4%±4.36% in 53 gestures identification of 10 subjects. Some classical machine learning algorithms such as support vector machine (SVM), linear discriminant analysis (LDA) and K Nearest Neighbor (KNN) are chosen to make comparison, where the SVM with Gaussian kernel function reaches to the maximum accuracy of 67.98%±4.56%. Two-way variance results show significant differences between each other. The experiment results show that the transfer learning is effective for decoding low-frequency sEMG for a large number of gestures.
Highlights
The cerebral cortex generates a movement command called motor intention as soon as the human body performs a certain action
By using some classical algorithms such as support vector machine (SVM) [4], linear discriminant analysis (LDA) [5], K Nearest Neighbor (KNN) [6] and random forests [7] etc., the satisfactory results can be obtained in surface electromyography (sEMG) pattern recognition [3], [8]
We find that there is no significant difference with the cases of single feature mean absolute value (MAV) and root mean square (RMS), except for the significant difference between the results of MRD and difference absolute standard deviation value (DASDV)
Summary
The cerebral cortex generates a movement command called motor intention as soon as the human body performs a certain action. Is very necessary to reduce the cost of information processing and the complexity of the classifier During this process, by using some classical algorithms such as support vector machine (SVM) [4], linear discriminant analysis (LDA) [5], K Nearest Neighbor (KNN) [6] and random forests [7] etc., the satisfactory results can be obtained in sEMG pattern recognition [3], [8]. Cote-Allard et al [25] performed a real-time study with transfer learning based on CNN They collected sEMG using an eight-channel Myo armband and controlled a 6-DoF robotic arm. Compared to the existing state of the art, the proposed method has the following advantages: 1) The number of classification actions is large for more natural and dexterous control, and the sampling frequency of sEMG is low for wearable low-cost application. 3) The suitable feature, data enhancement and network optimization are thoroughly discussed within the classification framework to improve the accuracy of classification
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call