Automatic Detection and Correction of Blink Artifacts in Single Channel Neural EEG Signals

Ocular Artifacts (OAs) are inevitable during Electroencephalogram (EEG) acquisition predominantly in human identified brain potentials, which makes the task of signal detection and correction is an essential and a critical function. This paper on one hand adopts an energy detection method to identify the artifacts and on the other hand performs wavelet thresholding in the identified zones to protect the neural information at non artifact regions. A multipronged combination of Wavelet Transform (WT) techniques, Discrete Wavelet Transform (DWT) and Stationary Wavelet Transform (SWT), and the threshold functions Universal Threshold (UT), Statistical Threshold (ST) and Hybrid Threshold (HT) are employed to identify the optimum combination for OAs removal. The performance of these methods at artifact regions is analytically assimilated with various standard metrics. Results of this study demonstrated that the SWT+HT method has yielded the desired outcome as compared to other methods of artifacts correction.