Hybrid HMM-MLP classifier for prosthetic arm control purpose

Author

Kwon, Jangwoo ; Shin, Chulkyu ; Lee, EungHyuk ; Han, Youngkwan ; Hong, SeungHong

Author_Institution

Dept. of Electron. Eng., Inha Univ., Inchon, South Korea

Volume

1

fYear

1996

fDate

26-29 Nov 1996

Firstpage

21

Abstract

This paper describes an approach for classifying electromyographic (EMG) signals using multilayer perceptrons (MLPs) and hidden Markov models (HMMs) hybrid classifier. Instead of using MLPs as probability generators for HMMs we propose to use MLPs as the second classifiers to increase the discrimination rates of myoelectric patterns. This strategy is proposed to overcome weak discrimination and to consider the dynamic properties of EMG signals. Three discrimination strategies (DHMM, cascaded DHMMs, and HMM with three subnet MLPs) for discriminating signals representative of 6 primitive class of motions are described and compared. The proposed strategy increase the discrimination results considerably. Results are presented to support this approach

Keywords

artificial limbs; backpropagation; electromyography; handicapped aids; hidden Markov models; medical signal processing; multilayer perceptrons; pattern classification; DHMM; EMG signals classification; MLP; backpropagation algorithm; cascaded DHMM; discrimination rates; dynamic properties; electromyographic signals; hidden Markov models; hybrid HMM-MLP classifier; multilayer perceptrons; myoelectric patterns; prosthetic arm control; signals discrimination; subnet MLP; Backpropagation algorithms; Band pass filters; Cost function; Electrodes; Electromyography; Equations; Hidden Markov models; Pattern matching; Prosthetics; Topology;

fLanguage

English

Publisher

ieee

Conference_Titel

TENCON '96. Proceedings., 1996 IEEE TENCON. Digital Signal Processing Applications

Conference_Location

Perth, WA

Print_ISBN

0-7803-3679-8

Type

conf

DOI

10.1109/TENCON.1996.608691

Filename

608691