Upper Limb Movement Recognition And Rehabilitation Application Based On SEMG Signal

Surface electromyography(sEMG)specifically and dispassionately reflects the movement of human muscles.As a helpful non-invasive EMG location strategy,it is broadly utilized within the field of human movement acknowledgment.In my nation,the number of individuals with upper appendage brokenness caused by maturing maladies,mischances and other components is expanding.The issue of utilizing upper limb restoration robots to assist patients with upper appendage restoration preparing has moreover pulled in far reaching consideration.Hence,human upper appendage movement design acknowledgment based on sEMG signals is utilized in related areas where recovery robots help patients in restoration preparing.This article primarily conducts design acknowledgment inquire about on the surface EMG flag produced by the human upper appendage hand development,and employments the existing restoration preparing stage to confirm the possibility of the surface EMG flag application in recovery preparing.The work of this paper has five parts: EMG flag collection,preprocessing,highlight extraction,design acknowledgment and classification and its application in restoration stage.The first part: the collection of sEMG.Firstly,analyze the era component of sEMG and the lower arm muscle gather included in hand development,and plan the flag collection test.Use the MYO armband to gather the surface EMG flag created by the human upper appendage hand development.The collected 6 sorts of hands The activity signals are record finger bowing,file finger expansion,center finger twisting,center finger expansion,ring finger bowing,ring finger expansion.The second part:pretreatment.Due to the commotion doped within the sEMG,the channel is utilized to expel the clamor exterior the recurrence run where the compelling data of sEMG is found,and after that the dynamic portion of the flag is identified,and the information parts created amid hand developments are sifted out,and the subjects are prohibited when they are resting.The created information part.The third part: include extraction.The three agent strategies of time space highlight,recurrence space include and time-frequency space highlight are presented and analyzed.Agreeing to the strategy of include assessment based on Davidson Forttin list,compelling highlights are chosen.The fourth part: design acknowledgment.The classification strategy of the bolster vector machine sort and the vector canonical bit work estimation strategy is utilized to classify and recognize the surface EMG flag,and the molecule swarm calculation and the blossom fertilization calculation are individually presented to the parameters that influence the classification execution of the back vector machine and the vector canonical bit work guess strategy The parameters are optimized.The acknowledgment rates of the two classifiers come to 80.56% and 94.44% individually.The five part:Experimental confirmation of restoration stage.After the hand development is discharged through the classifier,the upper computer sends a development command to the controller to form the recovery preparing stage move concurring to the acknowledgment result,and at long last it is concluded that the exactness of the recovery preparing platform's activity execution comes to more than 80%.The over outlines the achievability of sEMG-based movement acknowledgment for upper appendage recovery preparing.