期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2016
卷号：5
期号：7
页码：12081
DOI：10.15680/IJIRSET.2016.0507010
出版社：S&S Publications
摘要：EEG or Electroencephalography records the electrical activity occurring inside the brain due tocommunication that takes place between neurons which constitute the brain. It is a vital bio-potential that reflectsvariations in brain’s intricate internal activities and hence can provide valuable understanding of the brain’s working. Inrecent times it has been found to be useful in spotting neural ailments and also in the development of Human MachineInterfaces (HMI). However, EEG is highly vulnerable to being corrupted by non-cognitive signals whether biologicalor non-biological, called artifacts, which can considerably affect the end performance of an EEG based system. Severalapproaches have been suggested in literature to remove these artifacts-automatically as well as manually. In this paper anew system based on the combined use of ICA, DWT and Hierarchical Clustering is proposed along with theimplementation of two existing techniques based on DWT+ICA and ICA+HC. Technique-1 based on ICA and DWTefficiently removes non-physiological signals while the technique-2 based on ICA and Hierarchical Clusteringperforms moderately well for both types of noise signals. The proposed technique is found to be very effective inremoving biological artifacts and performs moderately well for non-biological signals.
