Attention-Enhanced Frequency-Split Convolution Block for sEMG Motion Classification: Experiments on Premier League and Ninapro Datasets

被引：2

作者：

Ergeneci, Mert ^{[1
,2
]}

Bayram, Erkan ^{[2
,3
]}

Binningsley, David ^{[4
]}

Carter, Daryl ^{[5
]}

Kosmas, Panagiotis ^{[1
]}

机构：

[1] Kings Coll London, Fac Nat Math & Engn Sci, Dept Engn, London WC2R 2LS, England

[2] Neurocess Ltd, London WC2A 2JR, England

[3] Univ Illinois, Dept Elect & Comp Engn, Coordinated Sci Lab, Urbana, IL 61801 USA

[4] Manchester United Football Club, Manchester M16 0RA, England

[5] Leeds United Football Club, Leeds LS11 0ES, England

来源：

IEEE SENSORS JOURNAL | 2024年 / 24卷 / 04期

关键词：

Sports; Adaptation models; Sensors; Feature extraction; Kernel; Muscles; Data models; Attention; convolutional block attention module (CBAM); convolutional neural network (CNN); gesture recognition; motion classification; neural networks; sports science; surface electromyography (sEMG); SURFACE EMG; PROFESSIONAL FOOTBALL; GESTURE RECOGNITION; INJURIES; RECOMMENDATIONS; EPIDEMIOLOGY; DIAGNOSIS; SIGNAL; TORQUE;

D O I：

10.1109/JSEN.2023.3345731

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This article presents convolutional octave-band zooming-in with depth-kernel attention learning (COZDAL), a versatile deep learning model designed for surface electromyography (sEMG) motion classification. Specifically focusing on sports movements involving the hamstring muscle, the model employs attention mechanisms across various frequency bands, kernel sizes, and hidden layer depths. The proposed method has been extensively evaluated on the benchmark Ninapro dataset and a custom soccer dataset. The results demonstrate substantial improvements over the existing state-of-the-art models, with an accuracy of 95.30% on Ninapro DB2, outperforming the previous best by 3.29%, and an accuracy of 98.80% on Ninapro DB2-B, an 8.66% enhancement. Remarkably, COZDAL exhibits a performance accuracy of 96.30% on a soccer dataset gathered from 45 elite-level athletes representing two clubs in the English Premier League (EPL). This result, achieved without parameter tuning, highlights the model's adaptability and exceptional efficacy across diverse motion scenarios, sensors, subjects, and muscle types.

引用

页码：4821 / 4830

页数：10