Comparison of Artifact Removal Techniques between DCT and DWT in Brain Signals

Abstract

The brain signal activity recorded from various sides of brain scalp is known as
Electroencephalogram (EEG). The artifacts are the disturbances in EEG and its removal techniques
are treated as a significant task among the researchers since last decade. In the proposed work the
authors have taken Discrete cosine transform (DCT) as an artifact removal technique and its results
are compared with the results of Discrete wavelet transform (DWT) and found that the DCT gives
better result than DWT and also DCT is more computationally efficient than DWT. Before applying
DCT the EEG signal undergoes through the preprocessing stages and artifact channel detection
methods. With the help of artifact detection methods, the artifact removal technique is not applied
unnecessarily to those channels not having artifact. The sample entropy and kurtosis used in the
proposed work as an artifact detection technique shows better result and is given in result section.