EEG Signals Preprocessing for BCI Systems Using Visual Evoked Potentials

In this work, we focus on brain stimulation using the VEP (Visual Evoked Potentials) principle. The brain is stimulated by stimulating videos that contain different categories of animals. By stimulating, we generate an electrical signal that we can measure with an EEG device. We selected the Ultracortex Mark IV, which is affordable for relatively good sensing quality. The acquired signal is further processed - we remove the DC offset from the signal using Fourier transform and then filter the signal with a high-pass and low-pass filter. In the first scenario, we try to see if similar or identical signals are emitted from the brain when different people are stimulated by the same video. The second scenario investigates whether one person can recognize different categories of animals. The similarity of the signals was detected by means of correlation coefficients. This would confirm the correctness of the visual stimulations, and we could create a system for non-verbal human-machine communication.