电子学报 ›› 2013, Vol. 41 ›› Issue (12): 2479-2486.DOI: 10.3969/j.issn.0372-2112.2013.12.025

• 科研通信 • 上一篇    下一篇

基于HHT和CSSD的多域融合自适应脑电特征提取方法

李明爱1, 崔燕1, 杨金福1, 郝冬梅2   

  1. 1. 北京工业大学电子信息与控制工程学院, 北京 100124;
    2. 北京工业大学生命科学与生物工程学院, 北京 100124
  • 收稿日期:2012-10-12 修回日期:2013-01-16 出版日期:2013-12-25 发布日期:2013-12-25
  • 作者简介:李明爱 女,1966年生,河南鹤壁人,2006年于北京工业大学获得博士学位,现为北京工业大学副教授、硕导,主要从事脑机接口、智能信息处理与模式识别等领域的研究. E-mail:limingai@bjut.edu.cn 崔 燕 女,1987年生,江苏盐城人,2010年获得南京工程学院学士学位,现为北京工业大学模式识别与智能系统专业硕士研究生,主要研究方向为脑机接口、信息处理与模式识别.
  • 基金资助:

    北京市教委项目面上项目(No.KM201110005005);国家自然科学基金(No.61201362);北京市自然科学基金(No.7132021,No.4112011);北京工业大学基础研究基金(No.X4002011201101)

An Adaptive Multi-Domain Fusion Feature Extraction with Method HHT and CSSD

LI Ming-ai1, CUI Yan1, YANG Jin-fu1, HAO Dong-mei2   

  1. 1. College of Electronic Information & Control Engineering, Beijing University of Technology, Beijing 100124, China;
    2. College of Life Science & Biological Engineering, Beijing University of Technology, Beijing 100124, China
  • Received:2012-10-12 Revised:2013-01-16 Online:2013-12-25 Published:2013-12-25

摘要: 为改善运动想象脑电信号特征提取的自适应性和实时性,提出一种基于希尔伯特-黄变换(HHT)与共空域子空间分解算法(CSSD)的特征提取方法(HCSSD).在对脑电信号进行预处理的基础上,定义一种相对距离准则优选脑电极组合;计算脑电的Hilbert瞬时能量谱和边际能量谱,以获取脑电的时-频特征,并基于CSSD提取其空域特征,采用串行特征融合策略得到脑电的时-频-空特征;设计学习矢量量化神经网络分类器,实现脑电数据分类.在训练集与测试集间隔一周且减少导联数量的情况下,基于HCSSD对左手小指和舌头的运动想象ECoG脑电数据的平均识别率为92%.实验结果表明:HCSSD在增强特征提取方法的自适应性、改善实时性的同时,提高了脑电信号识别率,为便携式BCI系统在康复领域的应用创造了条件.

关键词: 脑机接口, 运动想象, 希尔伯特-黄变换, 共空域子空间分解, 特征融合, 自适应

Abstract: The adaptivity and real-time performance of feature extraction method are crucial in brain-computer interface.Based on Hilbert-Huang transform (HHT) and common spatial subspace decomposition (CSSD) algorithm,a novel feature extraction method,denoted as HCSSD,was proposed.Firstly,the motor imagery electroencephalography (EEG)/ electrocorticography (ECoG) was preprocessed,and a relative distance criterion was defined to select the optimal combination of channels.Secondly,Hilbert instantaneous energy spectrum and marginal energy spectrum of EEG/ECoG were calculated to extract time feature and frequency feature respectively.Then CSSD was applied to extract spatial feature.Furthermore,serial feature fusion strategy was adopted to obtain time-frequency-spatial feature.Finally,learning vector quantization neural network was designed to classify the EEG/ECoG data.The average recognition accuracy was 92% for the left small finger and tongue motor imagery ECoG tasks.Experiment results show that HCSSD can enhance the adaptivity and real-time performance of feature extraction,with the recognition accuracy improved.This method provides a new idea for the application of portable BCI system in rehabilitation field.

Key words: brain-computer interface (BCI), motor imagery (MI), hilbert-huang transform (HHT), common spatial subspace decomposition (CSSD), feature fusion, adaptivity

中图分类号: