Publications by authors named "Huiqiao Lu"
Multi-modality deep forest for hand motion recognition via fusing sEMG and acceleration signals.
Int J Mach Learn Cybern
November 2022
Article Synopsis
- Bio-signal based hand motion recognition is crucial for human-machine interaction, especially in controlling prosthetic devices, but achieving high accuracy with multiple input modalities remains challenging.
- The study introduces a multi-modality deep forest (MMDF) framework that effectively combines surface electromyographic (sEMG) and acceleration (ACC) signals for better recognition of hand motions, utilizing a three-stage process for feature extraction, dimension reduction, and classification.
- Tests on the "Ninapro DB7" database demonstrate that the MMDF framework surpasses competitors in accuracy, confirming that ACC signals enhance sEMG performance and validating MMDF as a strong approach for integrating bio-signals in motion recognition.