Application of surface Emg signal on forearms for finger classification

This paper covers the design, implementation, and testing of the classifier for fingers classification. Data acquisition is done by using a Biopac MP35 hardware. Surface EMG signal (sEMG) can be recorded from the human body in a non-invasive. SEMG signals in this study were obtained from the four-channel electrodes placed around the forearm. The flexion consists of each fingers (thumb, index finger, middle finger, ring finger and little finger). From these data the classifier is made using a threshold value segmentation taken from each finger on each channel. The results of the classifier are 80% for the thumb, 87% for index finger for, 80% for the middle finger, 87% for ring finger and 74% for the little finger.