一种新的基于小波变换的EEG视觉伪信号修正方法

A New Method of EEG Ocular Artifacts Correction Based on Wavelet Transform

提出了一种在基准EEG信号和视觉信号未知条件下,基于小波阈值去噪的方法来修正EEG信号中出现的视觉伪信号(OA)的新方法。这种方法实现了对原始EEG信号进行平稳小波变换(SW T);对低频系数进行两次阈值去噪;对去噪后的信号进行重构。实验结果表明:这种新方法在基准EEG信号和视觉信号未知条件下能有效去除OA,同时适用于眨眼和眼球运动所产生的伪信号。通过不同方法对采集的信号处理后进行比较,说明该方法的有效性。

This paper gives a new method of electroencephalogram （EEG） ocular artifact （OA） correction based on stationary wavelet transform （SWT） in the case of unknowing uncontaminated baseline EEG and electrooculogy ram（EOG）. The proposed method for EEG OA correction based on SWT consists of the following steps： SWT is applied to the corrupted EEG; the threshold of the coefficients in the lower frequency bands is performed twice~ the de-noised signal is reconstructed. Simulation result indicates this new method can remove OA effectively and it suits for the artifacts produced by blinks and eye movements. Finally ,the availability of this method is proved by comparing with different methods.