Transactions of the Society of Instrument and Control Engineers
Online ISSN : 1883-8189
Print ISSN : 0453-4654
ISSN-L : 0453-4654
Paper
The Evaluation of the Realtime Hand Motion Recognition System That Analyzes Forearm SEMG Using Multiclass SVM
Masachika FUTAMATAKentaro NAGATAKazushige MAGATANI
Author information
JOURNAL FREE ACCESS

2014 Volume 50 Issue 1 Pages 37-43

Details
Abstract
Electromyogram (EMG) is a kind of biological signal that is generated because of excitement of muscle according to the motor instruction from a brain. We have been experimentally developing the real-time hand motion recognition system by using 4 channels forearm SEMG signals. In our system, in order to classify measured SEMG SVM (Support Vector Machine) that has higher discriminability is used. Usually SVM is used as a non-linear classifier. However, In the conventional system that we developed, we used a canonical discriminant analysis (CDA) method. CDA method is based on linear discriminant function, and it has shown good experimental results. So, we think linear SVM also has good separate performance. Than before, these discussions are not enough. Therefore, we developed real-time hand motion recognition system that uses linear SVM or non-linear SVM as a discriminant function, and we have compared the discriminant ability between linear SVM, non-linear SVM. In this report, we will describe about the results of this experiment.
Content from these authors
© 2014 The Society of Instrument and Control Engineers
Previous article Next article
feedback
Top