Human Motor Imagery based EEG Signals Classification using KNN classifier

Brain computer Interface (BCI) is a technique which is used for body movements by utilizing different Electroencephalography (EEG) signal classification technique. BCI technology is a future for person suffering from paralysis. In this work, we have collected EEG signal data to found out the frequencies domain features so that human can perform physical tasks. EEG data where acquired from 10 healthy human subjects at NITTTR Chandigarh. To determine efficient, reliable algorithms for upgrading the classification accuracy of motor imagery (MI) based EEG signal is mostly recommended for the improvement of BCI systems. The performance of each frequency domain features are compared to each other's. In this work, we proposed an approach to overcome some difficulty related to movement. By using KNN classifier, the results are shown with the most relevant accuracy. With the help of KNN algorithm under frequency-domain feature, the accuracy result is obtained.