Sitemap
A list of all the posts and pages found on the site. For you robots out there, there is an XML version available for digesting as well.
Pages
Posts
portfolio
publications
Toward reducing the effect of force variations on electromyography pattern recognition by Mel-frequency spectrum
Published in 2021 IEEE International Conference on Real-time Computing and Robotics (RCAR), 2021
In this paper, we proposed to use the feature of log-Mel-frequency spectrum (log-MFS) to enhance the performance of EMG-PR method in classifying arm movements of different force levels. The performances of the proposed feature was investigated and compared with those of the conventional TD features. And results showed that the log-MFS feature has strong robustness to against force variations.
Recommended citation: Liu Y, Tian L, Zheng Y, et al. Toward reducing the effect of force variations on electromyography pattern recognition by Mel-frequency spectrum[C]//2021 IEEE International Conference on Real-time Computing and Robotics (RCAR). IEEE, 2021: 309-313.
Download Paper
A Novel Motion Recognition Method Based on Force Myography of Dynamic Muscle Contractions
Published in Frontiers in Neuroscience, 2022
This study proposed a novel RSC-based motion recognition method. In this method, a novel piezoelectret force-sensitive sensor was used to record the RSC signals that were produced by dynamic pressure distributions on the skin surface. The work of this study would provide help to improve the robustness of motion pattern recognition, and the proposed RSC-based method may be a promising alternative approach for multifunctional prosthesis control.
Recommended citation: Li X, Zheng Y, Liu Y, et al. A novel motion recognition method based on force myography of dynamic muscle contractions[J]. Frontiers in Neuroscience, 2022, 15: 783539.
Download Paper
Simultaneous hand/wrist motion recognition and continuous grasp force estimation based on nonlinear spectral sEMG features for transradial amputees
Published in Biomedical Signal Processing and Control, 2023
In this paper, we proposed a framework of fusing the robust motion classification and continuous grasp force estimation. In the framework, three new spectral features were extracted from the nonlinear Melfrequency spectrum of EMG signals, in which the difference of spectrum power between adjacent channels (ch-MFSLD) and adjacent frequency bands (fr-MFSLD) were capable of strong robustness to force variations, and the feature of average power of each frequency band across channels (meanfr-MFSL) was used to estimate the continuous grasp force.
Recommended citation: Li X, Liu Y, Zhou X, et al. Simultaneous hand/wrist motion recognition and continuous grasp force estimation based on nonlinear spectral sEMG features for transradial amputees[J]. Biomedical Signal Processing and Control, 2023, 85: 105044.
Download Paper
A Novel Transformer-Based Approach for Simultaneous Recognition of Hand Movements and Force Levels in Amputees Using Flexible Ultrasound Transducers
Published in IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2023
In this paper, we used a self-designed lightweight, flexible, and wearable ultrasound transducer for data acquisition and proposed a novel Sonomyography Transformer (SMGT) model for simultaneously recognizing hand movements and force levels. This study may promote the applications of intelligent prosthetic hands and rehabilitation engineering.
Recommended citation: Peng X, Liu Y, Tan F, et al. A novel transformer-based approach for simultaneous recognition of hand movements and force levels in amputees using flexible ultrasound transducers[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2023, 31: 4580-4590.
Download Paper
A novel unsupervised dynamic feature domain adaptation strategy for cross-individual myoelectric gesture recognition
Published in Journal of Neural Engineering, 2024
In this paper, we proposed an optimized unsupervised feature domain adaptation strategy, namelyself- adaptive dimensional dynamic distributionadaptation (SD-DDA), which can automaticallyselect the optimal feature dimension based on theunlabeled target domain data from novel users toimprove the performance of cross-individual motionrecognition.
Recommended citation: Liu Y, Peng X, Tan Y, et al. A novel unsupervised dynamic feature domain adaptation strategy for cross-individual myoelectric gesture recognition[J]. Journal of Neural Engineering, 2024, 20(6): 066044.
Download Paper