Abstract:
Electromyography is a technique of recording neuromuscular signals from muscles during movement and rest position. Recorded signals are in the form of electric signals which can be easily analyzed and studied. s-emg signals are recorded from the surface of muscles. They can detect any kind of movement in muscles which can help diagnose between a healthy and a disable person. s-emg signal are analyzed using different technique of signals and systems. According to a specific movement their behavior and pattern is noted, and further actions can be taken. Analysis of s-emg is beneficial for mankind. Signal analysis techniques are used to make computerized wheelchairs, robotic arms and food cooking machines for disable persons. In this research, pattern recognition and machine learning technique are adopted to analyze signals. 7-time domain features are selected for pattern recognition. Signals are combined in a group and analyzed by LDA. 7 features make 6 groups with different combinations of features. Data from 10 movements of hand and from 5 subject were collected in SMME EMG lab. Results after computation and feature extraction shows that as we increase the no of features, accuracy is improved up to the combination of four features together. As further by five feature combination accuracy increases up to combination of seven altogether. ANOVA test is performed to finalize results. These result show there is a significant difference in the mean of seven groups and features combination
