Performance Assessment of EMG Pattern Recognition Techniques While Growing the Number of Movement Classes

DOI

No DOI

Published In

Volume 1, Issue 2 (2021)

Published Online

March-April 2021

Pages

09-13

Article Type

Research Article

Check for updates

No DOI available for cross-reference

In the past two or three significant length of assessment done in the field of myoelectric control, various experts have proposed a couple of models using a blend of different components and classifiers to extend the improvement classes, yet all that work fails to figure out if there is any association between's multi-class gathering and its accuracy. This paper revolves around finding the factors that finish up the imperative of improvement classes that AI computations can accuratelydifferentiateandtoevaluatetheperformanceofpatternclassificationtechniquesusingthesEMGsignal exactly when the amount of advancement classes is extended while keeping the straightforwardness of the structure. The results were gotten for eight channels sEMG signal using 7 extra energy space components and four rundown of capacities mixes more than 4 classifiers (Support Vector Machine(SVM), K-Nearest Neighbour(K-NN), Decision Tree(DT), and Naïve Bayes(NB)). Then, at that point, the amount of classes was extended in the method of 5, 7, 10, 12, and 15 classes to conclude the biggest number of improvement classes that the sEMG system .