A Multiscale Feature Extraction Network Based on Channel-Spatial Attention for Electromyographic Signal Classification

The applications of myoelectrical interfaces are majorly limited by the efficacy of decoding motion intent in the electromyographic (EMG) signal. Currently, EMG classification methods often rely substantially on handcrafted features or ignore key channel and interfeature information for classification tasks. To address these issues, a multiscale feature extraction network (MSFEnet) based on channel-spatial attention is proposed to decode the EMG signal for the task of gesture recognition classification. Specifically, we fuse the spatiotemporal characteristics of the EMG signal with different scales. Then, we construct the feature channel attention module and the feature-spatial attention module to capture more key channels features and more key spatial features. To evaluate the efficacy of the proposed method, extensive experiments are conducted on two public data sets: 1) NinaPro DB2 and 2) CapgMyo DB-a. An average accuracy of 86.21%, 90.77%, 92.53%, and 98.85% has been achieved in Exercises B, C, and D of NinaPro DB2 and CapgMyo DB-a, respectively. The experimental results demonstrate that MSFEnet is more capable of extracting temporal and spatial fused features. It performs well in generalization and has higher classification accuracy compared with the state-of-the-art methods.

Automated summary

This paper proposes a multiscale feature extraction network (MSFEnet) based on channel-spatial attention for decoding electromyographic (EMG) signals in gesture recognition classification tasks. By fusing the spatiotemporal characteristics and different scales of the EMG signal, feature channel attention module and feature-spatial attention module are constructed to capture more key channel and spatial features. Experimental results show that MSFEnet performs well in extracting temporal and spatial fused features, and achieves higher classification accuracy.