Multi-Class SVM Classification of Surface EMG Signal for Upper Limb Function

Electromyography (EMG) signal is electrical manifestation of neuromuscular activation, that provides access to physiological processes which cause the muscle to generate force and produce movement and allow us to interact with the world. In this paper, an identification of six degree of freedom for evaluating and recording physiologic properties of muscles of the forearm at rest and while contracting is presented. The first step of this method is to analyze the surface EMG signal from the subject´s forearm using wavelet packet transform and extract features using the singular value decomposition. In this way, a new feature space is generated from wavelet packet coefficients. The second step is to import the feature values into multi class support vector machine as a classifier, to identify six degree of freedom viz. open to close, close to open, supination, pronation, flexion and extension. The results showed that an accuracy of over 96% could be obtained for a six degree of freedom classification problem.