基于希尔伯特变换联合卷积神经网络的脑电信号识别方法

传统运动想象脑电信号识别方法需要人为提取大量特征,识别性能受研究人员经验影响较大,主观性强;提出一种基于希尔伯特变换(HT)联合卷积神经网络(CNN)的运动想象脑电信号自动识别方法,首先利用HT对原始EEG信号进行分析,实现一维数据向二维幅-相图像转换的同时增加信息提取维度;然后将其作为输入利用CNN层次化的对幅-相二维图像进行理解和解译,自动提取特征并完成分类识别,基于BCI竞赛中所用Graz数据集开展试验,结果表明相对于传统特征提取方法,文章所提算法在低、中、高信噪比条件下均能获得更好的识别性能,具有更强的噪声鲁棒性.

EEG signal recognition based on Hilbert transform and convolution neural network

Traditional recognition methods of motor imagery electroencephalogram (EEG) need to extract a lot of features artificially, and the recognition performance is greatly affected by the experience of researchers and has strong subjectivity. In this paper, an automatic recognition method of motor EEG signals based on deep learning is proposed. Firstly, the time-domain EEG is transformed into amplitude phase domain two-dimensional images by Hilbert transform, and then the convolution neural network is used to extract features and recognize different EEG signals. The experimental results based on Graz data set used in BCI competition show that the proposed algorithm can achieve better performance than the traditional feature extraction methods It has better recognition performance, and has higher robustness under the condition of low signal-to-noise ratio.