Description

Bioengineers continue to face a significant difficulty in the creation of prosthetic controls using neurophysiological inputs. Modern electromyography (EMG) continuous pattern recognition controllers are predicated on the dubious premise that repetitive muscle contractions result in reproducible steady-state EMG signal patterns. On the other hand, we provide a method for decoding wrist and hand motions by analyzing the signals that come just after the start of a contraction (i.e., the transient EMG). In our project collect the hand movements’ data for four actions (paper, scissor, and stone, normal) by using the EMG signals. Electromyography (EMG) measures muscle response or electrical activity in response to a nerve’s stimulation of the muscle. The test is used to help detect neuromuscular abnormalities. During the test, one or more small needles (also called electrodes) are inserted through the skin into the muscle. In our process use the .csv dataset for the signal classification and move on to the machine learning algorithms for the accuracy checking. Here use the three algorithms for the classification and also compare the accuracy result.