Movements Classification Based on Surface Electromyography Using Time-frequency Domain Features

Feature selection makes significant role in movement classification based on electromyography data. It is assumed that the efficiency of movement classification is improved when time-domain (TD) and frequency-time-domain (TFD) features are used together. To validate the hypothesis, classifiers based on Support Vector Machines (SVM), K Nearest Neighbors (KNN), Random Forest (RF) were trained using NinaPro DB5 dataset. Classification efficiencies of 91.3% and 92.8% were achieved for Recall and Precision metrics, respectively, using KNN. The proposed methods can be used to create human-machine interfaces using muscle activity data.