论文部分内容阅读
提出了一种应用离散小波变换(DWT)结合主分量分析(PCA)进行特征提取,然后用支持向量机(SVM)对P300进行分类的算法。该算法首先在一定预处理基础上使用离散小波变换对P300脑电信号分解,然后选取蕴含P300大多数信息的特征尺度进行小波重构,从而达到去噪增强的效果。然后使用PCA进行特征的提取和集中。最后使用支持向量机对提取到的特征分量进行分类。该算法将小波分解和主分量分析结合起来进行特征增强与提取,实验结果表明,该算法能够达到令人满意的正确分类率。
This paper proposes a method of feature extraction based on discrete wavelet transform (DWT) and principal component analysis (PCA), and then classifies P300 by Support Vector Machine (SVM). The algorithm firstly decomposes the P300 EEG by discrete wavelet transform based on a certain preprocessing, and then selects the feature scale containing the most information of P300 for wavelet reconstruction, so as to achieve the effect of denoising enhancement. Then use PCA for feature extraction and concentration. Finally, using the support vector machine to extract the feature components are classified. The algorithm combines wavelet decomposition and principal component analysis for feature enhancement and extraction. Experimental results show that the algorithm can achieve a satisfactory correct classification rate.