sEMG-based approach for estimating wrist and fingers joint angles using discrete wavelet transform

In this paper, we propose a support vector regression (SVR)-based approach for continuous estimation of wrist and fingers joint angles from surface Electromyography (sEMG) signals. In particular, the proposed approach utilizes the discrete wavelet transform (DWT) to generate a time-frequency representation of the sEMG signals. Then, we extract a set of features using the obtained DWT details and approximations coefficients. Finally, the computed features are used to train a set of SVR models that can estimate the wrist and fingers joint angles. In order to evaluate the performance of the proposed approach, we utilize a publicly available database, namely, the Ninapro database. Experimental results show that the proposed approach was able to achieve an average regression accuracy (r
²
) of 0.64.