Abstract

IBARRA-FUENTES, A.  and  MORALES-SANCHEZ, E.. Identification of 7 Movements of the Human Hand Using sEMG - 360° on the Forearm. Rev. mex. ing. bioméd [online]. 2021, vol.42, n.3, 1192.  Epub Mar 22, 2022. ISSN 2395-9126.  https://doi.org/10.17488/rmib.42.3.3.

This document shows the Identification of 7 gestures (movements) of the human hand from sEMG - 360° signals on the forearm. sEMG - 360° is the sEMG measurement through 8 channels every 45° making a total of 360°. When making a hand gesture, there will be 8 independent sEMG signals that will be used to identify the gesture. The 7 gestures to identify are: relaxed hand (closed), open hand (fingers extended), flexion and extension of the little finger, the ring finger, the middle finger, the index finger, and the thumb separately. One hundred samples for each gesture were captured and 3 feature extraction methods were applied in the time domain: mean absolute value (MAV), root mean square value (RMS) and area under the curve (AUC). A vector support machine (SVM) classifier was applied to each extractor. The gestures were identified and the percentage of accuracy in the identification was calculated for each extractor + SVM classifier using the confusion matrix method and including the 8 channels for each gesture. An accuracy of 99.52% was achieved for the identification of the 7 gestures applying sEMG - 360°.

Keywords : Electromyography; gesture; classifier.