Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (28): 128-130.DOI: 10.3778/j.issn.1002-8331.2009.28.038

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Feature extraction and classification study with energy entropy of IMFs to different mental tasks in EEG

LI Ying1,2,AI Ling-mei1,MA Miao1   

  1. 1.College of Computer Science,Shaanxi Normal University,Xi’an 710062,China
    2.Department of Computer and Information Engineering,Huainan Normal University,Huainan,Anhui 232038,China
  • Received:2009-03-24 Revised:2009-05-25 Online:2009-10-01 Published:2009-10-01
  • Contact: LI Ying

思维作业脑电的IMF能量熵特征提取与分类研究

李 营1,2,艾玲梅1,马 苗1   

  1. 1.陕西师范大学 计算机科学学院,西安 710062
    2.淮南师范学院 计算机与信息工程系,安徽 淮南 232038
  • 通讯作者: 李 营

Abstract: A new feature extraction and selection method based on the energy entropy of Intrinsic Mode Functions(IMFs) is presented.Three types of different mental tasks in EEG signals radiated from the targets are decomposed into their respective IMFs using the Empirical Mode Decomposition(EMD) procedure,and the energies of the same IMF of three types of signals are different.The energy entropies of the IMFs are calculated.K-neighbor classifier is used for classification experiments for three types of signals.The results show that the correct identification ratio of experiments above 75%.

Key words: Intrinsic Mode Function(IMF), EEG, Empirical Mode Decomposition(EMD), feature extraction, K Nearest Neighbors(KNN)

摘要: 提出了一种基于固有模态函数(Intrinsic Mode Function,IMF)能量熵的特征提取方法。对三类脑电思维信号分别进行了经验模态分解(Empirical Mode Decomposition,EMD),并得到与其相对应的IMF。试验发现对于不同类别的信号,同阶的IMF能量的判别熵有明显的不同。而采用K-近邻分类器对三类脑电信号进行了分类,发现基于最佳特征向量选择的分类试验的平均正确识别率达75%以上。

关键词: 固有模态函数, 脑电信号, 经验模态分解, 特征提取, K-近邻分类器

CLC Number: