Fig 1.
The sEMG gesture recognition overall framework based on MSFF-Net.
Fig 2.
A certain type of action EMG of subject 1.
Fig 3.
Z-score standardized comparison chart.
Fig 4.
Schematic diagram of sliding segmentation.
Fig 5.
The proposed MSFF-Net model.
Fig 6.
The network structure of the CBAM module with residuals, CSA, and SAS.
Table 1.
The CBAM module with residuals detailed network structure.
Table 2.
The Ninapro database summary table.
Fig 7.
The Ninapro database gesture action chart.
Fig 8.
The schematic diagram of stimulus and restimulus label.
Fig 9.
The enhanced comparison of time-warping data.
Table 3.
The classification results of DB2 and DB4 stimulus data and restimulus data.
Fig 10.
The average gesture recognition accuracy of DB2 subjects.
Table 4.
DB2 subject information sheet.
Fig 11.
DB2 subject identification accuracy and information bivariate plot.
Fig 12.
The recognition accuracy of 49 classes of movements of subjects.
Fig 13.
The average gesture recognition accuracy of DB4 subjects.
Table 5.
The BN layer sequence experiment.
Table 6.
The influence of early and late network weights on classification performance.
Table 7.
The classification and performance comparison of Ablation studies.
Table 8.
Comparison of the classification performance by the proposed network and other methods.