EMG Signal Classification Using Support Vector Discriminant Analysis

Classification of neuromuscular disorders using the intramuscular Electromyograph signals was obtained by improve the quality of the signal before feature extraction, and optimize the feature space to provide better discrimination ability. The signal quality evaluation was considered based on determining the best wavelet function for Electromyograph identification and denoising. Optimizing the feature space was performed based on the supervised feature projection method, the Support Vector Discriminant Aanalysis.