Multi-domain feature extraction from surface EMG signals using nonnegative tensor factorization

To propose a multi-domain feature extraction method of surface EMG signals is of great significance to EMG-based human-computer interface (HCI). In this paper, nonnegative Tucker decomposition (NTD)-one model of nonnegative tensor factorization (NTF)-is used to extract multidomain features of sEMG signals for classification. In the first step the sEMG data are transformed into multidimensional information using continuous wavelet transform and the 4-D sEMG tensor is established. Then the tensor is decomposed into four components (spatial components, spectral components, temporal components and category components) and the core tensor is the feature extracted. The feature after being eliminated and compressed are fed into KNN, LDA and SVM classifiers for the identification of condition classification. An experiment about elbow movements of 10 healthy participants was carried out to verify the validity of this algorithm. The result implied that NTF is a meaningful and valuable multidomain feature extraction method to EMG-based HCIs.