文献类型：期刊文献

英文题名：Pattern Recognition Of Hand Movements Based on Multi-Channel Mechanomyography In The Condition of One-Time Collection and Sensors Doffing and Donning

作者：Zhang, Yue[1]; Xia, Chunming[2,3]; Cao, Gangsheng[2]; Tongtong, Zhao[2]; Zhao, Yinping[4]

机构：[1] School of Mechanical Engineering, Nantong University, Nantong, 226019, China; [2] School of Mechanical and Power Engineering, East China University of Science and Technology, Shanghai, 200237, China; [3] School of Mechanical and Automotive Engineering, Shanghai University of Engineering Science, Shanghai, 201620, China; [4] School of Software, Northwestern Polytechnical University, Xi’an, 710072, China

年份：2023

外文期刊名：SSRN

收录：EI(收录号：20230067822)

语种：英文

外文关键词：Classification (of information) - Palmprint recognition - Support vector machines - Wearable sensors

摘要：Pattern classification of hand movements based on mechanomyography (MMG) has application value in the development of human machine interaction and wearable devices. In the condition of one-time collection, high classification accuracy can be acquired, however sensors doffing and donning is unavoidably to change the site and contact pressure of sensors, making harmful effect on classification accuracy. In the condition of sensors doffing and donning, eight-channel MMG of the forearms from subjects when they were performing four classes of hand movements were collected for 12 days, and pattern recognition of hand movements were investigate for one-time collection and sensors doffing and donning. After feature extraction and combination of feature subsets, broad learning system (BLS) was introduced to pattern recognition of hand movements based on MMG, compared with three other algorithms. In the condition of one-time collection, recognition rates of each class is higher than 99%, better than that by using support vector machine (SVM), which demonstrates the excellent learning ability of BLS. In the condition of sensors doffing and donning, the classification accuracy by using SVM is higher than that by using BLS. When using SVM, BLS or extreme learning machine, the classification accuracy increases when adding the number of data subsets gradually, which illustrates that increasing training data can alleviate the negative effect on classification accuracy caused by sensor doffing donning. ? 2023, The Authors. All rights reserved.