Regresar

An experimental comparative analysis among different classifiers applied to identify hand movements based on sEMG

Abstract:

This paper presents a comparative analysis among different methods of classifiers such as: Feedforward Neural Networks (FFN), Support Vector Machines (SVM), Naïve Bayes Classifier (NBC) and Linear Discriminant Analysis (LDA) applied for pattern recognition on electromyography signals (EMG) of forearm muscles to identify hand's movements. Movements to be recognized are: closed hand, open hand, hand flexed inwards, hand flexed out and relax position. Signals are obtained from a 'Myo Armband' device that has 8 dry sensors, from which a set of features is extracted: Mean Absolute Value (MAV), Root Mean Square (RMS), Variance (VAR) and Standard Deviation (STD). The procedure consists of two stages, first, training and validation, and second, focused on testing the performance of classifiers with test subjects. Finally, the best classification method is presented through the experimental accuracy measurement.