Reconstruction of incomplete surface electromyography based on an adversarial autoencoder network

Surface electromyography (sEMG) signals are often incomplete due to interferences during data measurement, which can degrade sEMG-based applications. To address this issue, this paper proposes a novel adversarial autoencoder model, called the SGMD-AAE, which includes a self-mask generator and a mult...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Biomedical signal processing and control Ročník 86; s. 105084
Hlavní autoři: Zou, Yongxiang, Cheng, Long, Han, Lijun
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 01.09.2023
Témata:
Adversarial autoencoder
Missing sEMG
Multi-view discriminator
Self-mask
Signal reconstruction
Missing sEMG
Signal reconstruction
Adversarial autoencoder
Self-mask
Multi-view discriminator
ISSN:1746-8094, 1746-8108
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Surface electromyography (sEMG) signals are often incomplete due to interferences during data measurement, which can degrade sEMG-based applications. To address this issue, this paper proposes a novel adversarial autoencoder model, called the SGMD-AAE, which includes a self-mask generator and a multi-view discriminator. The generator’s binary mask is replaced with a self-mask mechanism, and an adversarial loss is added to promote the reconstruction performance. The multi-view discriminator extracts and fuses deep features of sEMG from time and frequency domains to enhance the generator’s reconstruction ability. The SGMD-AAE model is evaluated on the benchmark NinaPro DB2 database, and the experimental results show that it significantly outperforms incomplete sEMG signals, reducing NRMSE by 88.04% and increasing PSNR by 116.21%. The proposed model also achieves high recognition accuracy for hand gesture recognition even in extreme cases where 90% of the sEMG signals are missing, with an average accuracy exceeding 84%. The effectiveness of the SGMD-AAE model is further verified on a self-collected dataset, demonstrating similar recognition results. •This paper proposed a new model, named SGMD-AAE, which provides a universal method to solve various signal missing problems caused by different reasons, such as poor communication, electrodes disconnection, and downsampling.•Experimental validation has been performed to assess the reconstructed sEMG signal performance. The outcomes indicate that the classification accuracy can exceed 84%, even when up to 90% of the sEMG signals are lost.•The experimental result indicates that the proposed SGMD-AAE model can reconstruct incomplete sEMG signals effectively, enabling the use of fewer electrodes in myoelectric control. This tool reduces computing resource consumption and is especially useful in resource-constrained scenarios.
ISSN:1746-8094
1746-8108
DOI:10.1016/j.bspc.2023.105084