Extração de características e classificação de sinais sEMG aplicados a uma prótese de mão virtual

Abstract

This work proposes the classification of motor tasks, using surface electromyography (sEMG) to control a prosthetic hand for rehabilitation of amputees. Two types of classifiers are compared: k-Nearest Neighbor (k-NN) and Bayesian (Discriminant Analysis). Motor tasks are divided into four groups correlated. The volunteers were healthy people (without amputation) and several analyzes of each of the signals were conducted. For offline analysis, the features used were: RMS (Root Mean Squared), VAR (Variance) and WL (Waveform Length). For online experimentation, it involved the use of feature of Discriminant of Bi-spectral. In both cases, either online or offline techniques were used to sliding windows. A model is proposed for reclassification using cross-validation in order to validate the classification, and a visualization in Sammon Maps is provided in order to observe the separation of the classes for each set of motor tasks. The proposed method can be implemented in a computer interface providing a visual feedback through an artificial hand prosthetic developed in Visual C++ and MATLAB commands