ABSTRACT
Electromyography has been extensively used in a variety of fields. By using feature extraction to detect and analyze the surface EMG signal of Electromyography, muscle fatigue caused by daily life workout could be detected more timely. Here we intend to utilize this feature of using feature extraction on electromyography to offer professional advice for at home work out due to the deduction of outing caused by COVID-19. In this work, multiple time window (MTW) features have been used to distinguish the surface electromyography (sEMG) signals between muscle fatigue during arm movements by using Python. The sEMG signals are monitored from the biceps muscle of 3 healthy subjects. 4 window functions named boxcar function, hamming function, blackman function, and kaiser function and 24 features are extracted. 4 classifiers named Decision Tree, Random Forest, Support Vector Machine, and Naive Bayes are used in this research. The classifier using MTW features compared with the classifier without MTW feature. The Random Forest classifier has the greatest accuracy of 95.16%.