Abstract:
Electromyographic signals have a considerable importance in robotic hand prosthesis and various biomedical applications. The analysis of these signals for pattern recognition of arm movements is helpful to facilitate the handicap individuals with upper limb impairment or paralyzed individuals who are able to reinstitute innate control of hand. The techniques based on this analysis are not natural and demands prolonged training duration but the application of these methods is emerging successfully. As the difference in the configuration of these signals depends on the different muscle activities, they need to be recorded from the patients hand with the help of electrodes which may be contaminated with noises or undesired signals which affect the output accuracy. To ensure the detection of data with reduced noise and to execute the optimal performance from the analysis, the signals are preprocessed. The data collected for the 52 movements from 27 different subjects is provided by NinaPro database which is viable in attaining effective hand prosthetics and allowing the whole research community to add more advancement to this field. The purpose of this thesis is to analyze the dataset from this easily accessible database for twelve finger and hand movements acquired from 27 subjects. These signals are then processed with frequency filtering and data windowing to make them convenient for further use. Features extraction module consisting of four different features was then computed for these signals which are crucial step in gaining more dexterous myoelectric control of hand prosthesis. This processed data was then tested for two different classifiers to examine their percent classification accuracy. Two classifiers selected for this purpose were Linear Discriminant Analysis classifier and Artificial Neural Network classifier. The data classified with Linear Discriminant Analysis gives the mean classification accuracy of 85.41% while Artificial Neural Network classifier shows 91.14%. These performance results revealed that Artificial Neural Network performs better in the classification and recognition of data for hand movements as compared to Linear Discriminant Analysis.
