Signal Hybrid HMM-GA-MLP Classifier for Continuous EMG Classification Purpose

초록

This paper describes an approach for classifying electromyographic signals using a multilayer perceptron with genetic algorithm and hidden Markov models hybrid classifier. Instead of using MLP as probability generators for HMM's we propose to use MLP with GA as the second classifiers to increase discrimination rates of myoelectric patterns. The GA for MLP was driven to boost the learning time when it applied to backpropagation algorithm. This strategy is proposed to overcome weak discrimination and to consider dynamic properties of EMG signals. Four discrimination strategies for discriminating signals representative of 6 primutuve class of motions are described and compared. The proposes strategy increase the discrimination results considerably results are presented to support this approach.