基于离散小波变换提取脑机接口中脑电特征

在脑机接口中，针对脑电特征提取利用单一种类信息、使用数据量大、分类性能较差等缺点，提出一种新颖的基于离散小波变换的方法。分析了小波变换特征提取的特点和特征表示方式，用Daubechies类db4小波函数对脑电信号进行6层分解，抽取小波变换各子带关键的部分逼近系数、小波系数、小波子带系数均值组成特征向量。以分类正确率为指标检验了提取特征的性能。实验结果表明，这种方法能够利用少量数据提取脑电信号本质特征，具有较高的分类性能，为利用脑电识别人的不同意图提供了快速而有效的手段。

Extracting EEG Feature in Brain-computer Interface Based on Discrete Wavelet Transform

In brain-computer interfaces(BCIs),a novel method of extracting electroencephalography(EEG) features based on discrete wavelet transform(DWT) was proposed. The method aims at solving problems such as single information is used,large amount of data are required and the classification performance is low in BCIs.The characteristic of extracting features with wavelet transform and some representation manners of features were analyzed.The EEG signal was decomposed to six levels by Daubechies db4 wavelet function.A part of key approximation coefficients,wavelet coefficients and coefficient averages of each wavelet subband were extracted and formed a feature vector.The classification performance of extracted features was tested by an index called correct classification rate.The experiment results showed that the proposed method could extract substantial features using fewer data and displayed high classification performance.The method offers a fast and effective measure for recognizing human different intentions by EEG.