Sequential recognition of EMG signals using bayes-optimal feature extraction - Application to the control of bio-prosthetic hand

Abstract

The paper presents a concept of bio-prosthesis control via recognition of user intent on the basis of miopotentials acquired of his body. The contextual (sequential) recognition is considered in which the Bayes-optimal feature extraction from EMG signal is applied. An experimental comparative analysis of the proposed sequential classification algorithms and feature extraction procedures for real data is performed and results are discussed.