A Modified Digital Image Processing Enabled Stroke Rehabilitation System based on Wavelet Transform

Abstract

stroke rehabilitation therapy is for people suffering from paralysis. Regular procedures need the involvement of analyst for the duration of the plenary, requires excessive amount. Recently, many approaches have been proposed with control strategy through gesture recognition. The main objective of this work is hand gesture control for stroke rehabilitation based on dwt based feature extraction method. Establishment of artificial machine based computer system assistance provides instruction for paralysis recovery system. The proposed system uses reverse biorthogonal wavelet with two-level decomposition. Preprocessing and image segmentation of real time movement of the hand gestures. Feature extraction is done using discrete wavelet transform (dwt) based approach which has very flexible and adaptable even at the cost of imperfect reconstruction. The prosthetic based robotic hand for human finger movement recovery function. Human being finger movement process helps to rotate motors packed inside the robotic finger section. Therefore, the five finger machine finger movement can imitate the real time human being gesture in a regular manner, which indicates the newly modified machine represents as a education device to provide recovery for the persons affected after paralysis.