脑电信号(EEG)是脑神经细胞的电生理活动在大脑皮层的反映,但采集到的脑电信号一般都含有大量噪声。为了保留有效信息同时消除尽可能多的噪声,提出通过构造虚拟通道将集合经验模态分解与独立成分分析相结合的脑电信号去噪方法。首先,对脑电信号进行EEMD分解得到固有模态函数(IMF)分量,根据相关性准则筛选含噪声成分多的IMF分量构造虚拟通道进行ICA去噪;然后,将消噪后的结果与含信号成分多的IMF分量进行重构再次ICA去噪,得到最终去噪信号。为了验证EEMD-ICA去噪方法的有效性,以信噪比、均方根误差作为评价指标,将该方法与小波去噪法、EEMD去噪法、ICA去噪法进行比较,结果表明,EEMD-ICA去噪后的信噪比高于其它方法,均方根误差小于其它方法,综合分析该方法能更好地消除噪声。
Abstract
The electroencephalograph (EEG) signal is the electro physiological activity of brain cells, it is reflected in the scalp surface. However it is usually interfered by noises during signal acquisition process. In order to reserve the effective information and eliminate as much noise as possible, a method of ensemble empirical mode decomposition (EEMD) combined with independent component analysis (ICA) is introduced. Firstly, the EEMD decomposition can get a certain number of intrinsic mode function (IMF) of EEG signals. The virtual channel is reconstructed by the IMF components with more noise components which are selected based on correlation coefficient and de-noised by ICA algorithm. Secondly, the denoised results and the IMF components with multiple signal are reconstructed. Finally, the reconstructed signal is denoised by ICA again, and the final denoised signal is obtained. The experimental results show that the mentioned method has better signal-to-noise ratio and smaller RMSE than the other denoising methods, including wavelet denoising, EEMD denoising and ICA denoising. It shows that the mentioned method can denoise better.
关键词
计量学 /
脑电信号 /
去噪 /
集合经验模态分解 /
独立成分分析 /
虚拟通道
Key words
metrology /
EEG signal /
denoise /
ensemble empirical mode decomposition /
independent component analysis /
virtual channel
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
［1］范金锋, 邵晨曦, 王剑, 等. 醉酒者脑电和正常脑电非线性特性的比较评估 ［J］. 中国生物医学工程学报, 2008, 27(1): 18-28.
Fan J F, Shao C X, Wang J, et al. A Comparative Evaluation of Nonlinear Characteristics of EEG for Alcoholic and Normal People ［J］. Chinese Journal of Biomedical Engineering, 2008, 27(1): 18-28.
［2］王天伟, 张厚江, 路敦民, 等. 基于小波去噪和经验模态分解算法的足尺人造板动态称重研究 ［J］. 计量学报, 2017, 38(3): 300-303.
Wang T W, Zhang H J, Lu D M, et al. Dynamic Weighing of Full-size Wood Composite Panels Based on Wavelet Denoising and Empirical Mode Decomposition Algorithm ［J］. Acta Metrologica Sinica, 2017, 38(3): 300-303.
［3］王志芳,王书涛,王贵川, 等. 基于小波优化EEMD的二氧化硫检测［J］. 计量学报, 2020, 41(6): 752-758.
Wang Z F,Wang S T, Wang G C, et al. Detection of SO2 Based on EEMD Optimized by Wavelet［J］. Acta Metrologica Sinica, 2020, 41(6): 752-758.
［4］柏文琦, 李林峰, 陈红江, 等. EMD和SVM在变速器故障诊断中的应用 ［J］. 计量技术, 2015, (5): 3-6, 13.
Bai W Q, Li L F, Chen H J, et al. Application of EMD and SVM in transmission fault diagnosis ［J］. Measurement Technique, 2015, (5): 3-6, 13.
［5］罗志增, 袁飞龙, 高云园. 小波和希尔伯特变换在脑电信号消噪中的对比研究 ［J］. 计量学报, 2013, 34(6): 567-572.
Luo Z Z, Yuan F L, Gao Y Y. The Comparative Research on Wavelet and Hilbert Transform in the EEG Denoising ［J］. Acta Metrologica Sinica, 2013, 34(6): 567-572.
［6］余炜, 韩强, 马晶晶, 等. 基于EMD和SVM的脑电信号处理方法 ［J］. 昆明理工大学学报(自然科学版), 2012, 37(6): 38-42.
Yu W, Han Q, Ma J J, et al. EEG Signal Processing Method Based on EMD and SVM ［J］. Journal of Kunming University of Science and Technology (Natur-al Science Edition), 2012, 37(6): 38-42.
［7］刘彬, 蒋金水, 于伟凯. EMD相关度去噪及其在轧机信号处理中的应用 ［J］. 计量学报, 2009, 30(1): 73-77.
Liu B, Jiang J S, Yu W K. EMD Denoising Method and Its Application in the Processing for Rolling Mill Signal ［J］. Acta Metrologica Sinica, 2009, 30(1): 73-77.
［8］孟明, 杨国雨, 高云园, 等. 基于EEMD与DSS-Apen的脑电信号消噪方法 ［J］. 传感技术学报, 2018, 31(10): 1539-1546.
Meng M, Yang G Y, Gao Y Y, et al. EEG De-Noising Method Based on EEMD and DSS-Apen ［J］. Chinese Journal of Sensors and Actuators, 2018, 31(10): 1539-1546.
［9］朱江淼,闫迪,高源,等. 基于改进EEMD和LSSVM的单频周跳探测与修复方法［J］. 计量学报, 2019, 40(3): 491-497.
Zhu J M, Yan D, Gao Y,et al. The Method for Single-frequency Cycle-slip Detection and Reparation Based on Modified EEMD and LSSVM［J］. Acta Metrologica Sinica, 2019, 40(3): 491-497.
［10］韩刚, 张建文, 褚鑫. 基于EEMD自适应阈值去噪的电能质量扰动检测与定位研究 ［J］. 电测与仪表, 2014, 51(2): 45-49.
Han G, Zhang J W, Chu X. Research on Denoising of Power Quality Disturbance Detection and Location Based on EEMD Adaptive Thresholding ［J］. Electrical Measurement & Instrumentation, 2014, 51(2): 45-49.
［11］Tran Y, Craig A, Boord P, et al. Using independent component analysis to remove artifact from electroencep-halographic measured during stuttered speech ［J］. Medical Biological Engineering Computing, 2000, 42 (5): 627-633.
［12］秦娜, 金炜东, 黄进, 等. 基于EEMD的高速列车转向架故障诊断 ［J］. 计算机工程, 2013, 39(12): 1-4, 10.
Qin N, Jin W D, Huang J, et al. Fault Diagnosis of High Speed Train Bogie Based on Ensemble Empirical Mode Decomposition ［J］. Computer Engineering, 2013, 39(12): 1-4, 10.
［13］杨默涵, 陈万忠, 李明阳. 基于总体经验模态分解的多类特征的运动想象脑电识别方法研究 ［J］. 自动化学报, 2017, 43(5): 743-752.
Yang M H, Chen W Z, Li M Y. Multiple Feature Extraction Based on Ensemble Empirical Mode Decom position for Motor Imagery EEG Recognition Tasks ［J］. Acta Autom Atica Sinica, 2017, 43(5): 743-752.
［14］邱天爽, 武建. 多参考信号的ICA方法及其在去除脑电伪差中应用 ［J］. 大连理工大学学报, 2008, 48(5): 740-743.
Qiu T S, Wu J. ICA Method of Multi-Reference Signal and Its Application in Removing EEG Pseudo ［J］. Journal of Dalian University of Technology, 2008, 48(5): 740-743.
［15］杨杰, 陈捷, 洪荣晶, 等. 基于EEMD-多尺度主元分析的回转支承信号降噪方法研究 ［J］. 中南大学学报(自然科学版), 2016, 47(4): 1173-1180.
Yang J, Chen J, Hong R J, et al. Electroencephalogram Denoising Method Based on EEMD and Improved Lifting Wavelet ［J］. Journal of Central South University of Technology(Natural Science Edition), 2016, 47(4): 1173-1180.
［16］孟明, 鲁少娜, 马玉良. 基于EEMD与改进提升小波的脑电信号消噪方法 ［J］. 计算机工程, 2016, 42(4): 313-317.
Meng M, Lu S N, Ma Y L. Independent Componenta Analysis and Its Application In Signal Processing ［J］. Computer Engineering, 2016, 42(4): 313-317.
基金
国家自然科学基金(61201111);燕山大学博士基金(BL17026)
PDF(2504 KB)
