基于EEMD和模糊阈值的去噪方法

为了提高EEMD分解中噪声主导模态的去噪效果,利用模糊隶属度的优势,提出了一种EEMD和模糊阈值相结合的去噪方法。首先用二范数计算各个本征模态函数(IMF)与观测信号的概率密度函数(PDF)之间的相似度,得到噪声主导的IMF;然后对噪声主导的IMF进行模糊阈值处理,以去除IMF中的噪声;最后将所有的IMF重构得到消噪信号。分别采用仿真信号和ECG信号进行去噪实验,结果均表明,所提方法的去噪效果整体上优于小波半软阈值方法和基于EMD的间隔阈值(EMD-IT)方法。     

基于夹角余弦和模糊阈值的EEMD去噪方法

针对非线性非平稳信号的去噪问题,结合EEMD分解信号的自适应特性,提出一种基于夹角余弦和模糊阈值的去噪方法。首先用夹角余弦法计算各个本征模态函数（IMF）与观测信号之间的相似度,以相似度曲线的首个极小值的后一个位置为分界点将分解出的IMF分为噪声主导模态和信号主导模态;然后根据VisuShrink阈值易“过扼杀”细节系数和SUREShrink阈值易“过保留”噪声系数的特点,利用模糊阈值对噪声主导的IMF进行处理;最后将所有的IMF重构得到消噪信号。分别采用仿真信号和真实ECG信号进行去噪实验。结果表明,所提方法在整体性能上优于小波半软阈值方法、基于EMD的软阈值（EMD-Soft）和间隔阈值（EMD-IT）方法,是一种有效的去噪方法。     

基于总体平均经验模态分解的语音增强算法研究

总体平均经验模态分解EEMD(Ensemble Empirical Mode Decomposition)虽然能够在一定程度上抑制模态混淆,但添加的白噪声不能被完全中和,对所有本征模态函数IMF(Intrinsic Mode Function)分量进行集成平均等增加了计算工作量。基于EEMD和结合小波阈值去噪思想,提出改进的EEMD方法。首先对原始信号进行EEMD分解,得到一系列IMF分量;其次对筛选后的每个IMF计算噪声强度;然后采用小波启发式阈值估计噪声并计算阈值;最后以软阈值的方式滤除每个IMF中噪声并重构信号还原出增强的语音。通过分析仿真信号和实测信号,结果表明:该算法对带噪语音有很好的滤波效果,与其他同类算法相比提高信噪比2~4 d B。     

基于CEEMD阈值和相关系数原理的MEMS陀螺信号去噪方法

针对经验模态分解（EMD）方法易产生模态混叠问题,而集成经验模态分解（EEMD）方法又存在重构误差较大的缺陷,提出了一种基于完备集成经验模态分解（CEEMD）阈值滤波和相关系数原理的MEMS陀螺信号去噪方法。首先通过CEEMD方法对陀螺信号进行有效完备的分解,并利用相关系数原理合理确定分解后噪声分量与有效分量的界限。在此基础上,通过借鉴小波阈值处理方式和EMD阈值设置方法,对信号进行阈值滤波去噪。对仿真信号和实际MEMS陀螺信号的研究结果表明,CEEMD阈值去噪方法的去噪效果要优于CEEMD、EEMD、EMD强制去噪方法和小波分析方法。这也充分体现了其在MEMS陀螺信号去噪应用中的可行性和有效性。     

基于ICA算法的集合经验模态分解去噪方法

针对非线性非平稳信号的去噪问题,提出了一种基于独立分量分析(Decomposition Components Analysis,简称ICA)算法的集合经验模态分解去噪方法。首先利用白噪声辅助数据分析方法——集合经验模态分解(Ensemble Empirical Mode Decomposition,简称EEMD)有效的抑制了经验模态分解(Empirical Mode Decomposition,简称EMD)中存在的端点效应和模态混叠现象,然后利用ICA算法对含噪信号经过EEMD分解后的有限个固有模态函数(Intrinsic Mode Function,简称IMF)进行去噪处理,有效的分离出若干个有效的语音信号分量,并对其进行语音重构,最后与小波阈值去噪方法进行比较,通过仿真可以看出,该方法对于信号去噪较为理想。     

基于改进EEMD的心电信号基线漂移消除方法

针对传统方法滤波效果不佳的问题,本文提出了基于改进集合经验模态分解（Ensemble empirical mode decomposition,EEMD）的消除心电信号基线漂移方法。该方法克服了经验模态分解（Empirical mode decomposition,EMD）模态混叠的问题,并对EEMD方法存在的问题和不足进行改进,建立集合经验模态分解方法中加入辅助白噪声大小的可依据准则,从而确定加入的辅助白噪声大小以及集合平均次数这两个重要参数。它从含噪心电信号中提取基线漂移信号,然后重构其余本征模函数（Intrinsic mode function,IMF）分量得到"干净"的心电信号,为后续的研究提供前提。经实验验证表明：相较于传统方法,这种方法能够提高信噪比、降低均方差、保持特征波形、去噪更加彻底,很好地解决了心电信号低频成分损失的问题。     

基于EMD阈值算法的脉冲涡流信号消噪

针对脉冲涡流信号夹杂着较多的高频噪声,提出了一种新的经验模态分解阈值消噪算法。首先将信号分解为多个本征模态函数（Intrinsic Mode Function,IMF）,对信噪比低的高频IMF进行减小噪声能量处理后得到重组信号;再对重组信号进行EMD分解后根据白噪声统计特性对IMF筛选,对噪声含量多的IMF进行小波阈值消噪;最后将处理过的IMF与噪声含量少的IMF重构得到消噪后的信号。实验仿真的结果和数据表明,该方法可以减少失真,获得更高的信噪比,能够较好地消除噪声的干扰恢复出原始的信号。     

基于ＥＥＭＤ域统计模型的话音激活检测算法

提出了一种基于EEMD域统计模型的话音激活检测算法。算法首先利用总体平均经验模态分解(Ensemble empirical mode decomposition,EEMD)对带噪语音进行分解,得到信号的本征模式函数(Intrinsicmode function,IMF)分量,选择与原信号的相关性最高的两个分量相加组成主分量;然后对主分量进行频域分解,引入统计模型,求出EEMD域特征参数;最后利用噪声与语音的EEMD域特征参数的不同来进行语音激活检测。实验结果表明,在不同信噪比情况下,本文算法性能优于目前常用的VAD算法,特别在噪声强度大时体现出明显的优势。     

优化VMD与NLM结合的信号去噪

为解决人体心电信号易出现噪声干扰的情况,提出一种自选取参数的变分模态分解(variational mode decomposition)与非局部均值(no.local means) ECG信号去噪方法.针对人为选取变分模态分解参数不准确的问题,运用一种自适应的参数确定方法,根据不同的原始信号,确定不同的参数并对信号进行分解,得到一组信号分量,计算各个分量的样本熵,根据样本熵的值,选取出噪声主导分量和有效分量,对噪声主导信号进行NLM去噪,将去噪后的信号分量与剩余的有效信号分量进行重构得到去噪信号.实验结果表明,改进参数的VMD-NLM算法相较其它算法,去噪信号的信噪比(SNR)和均方误差(MSE)最好,对人体心电信号的去噪效果最为优秀.     

基于EEMD和WT的运动想象脑电信号消噪方法

采集到的运动想象脑电信号MI EEG(Motor Imagery Electroencephalogram)通常含有大量噪声信号.为了消除噪声同时保留尽可能多的有效信号,本文提出了将集合经验模态分解EEMD(Ensemble Empirical Mode Decomposition)与改进小波阈值法相结合的消噪方法.改进小波阈值法采用了新的阈值选取规则和阈值函数.首先对信号进行EEMD分解,然后再对高频固有模态函数IMF(Intrinsic Mode Functions)进行改进小波阈值处理,最后将处理后的高频IMF分量和低频IMF分量进行重构得到消噪信号.以信噪比和均方根误差作为消噪效果的定量评价指标,将本文提出的方法与单纯使用EEMD分解消噪法、单独使用改进小波阈值消噪法、EMD与改进小波阈值法相结合消噪法进行比较,结果表明,本文提出的消噪法优于其他三种消噪法.     

基于改进小波阈值-CEEMDAN算法的ECG信号去噪研究

Electrocardiogram (ECG) signal denoising has always been a hot research issue. In order to eliminate the noises in ECG signal, a denoising method based on adaptive complete set empirical mode decomposition (CEEMDAN) and wavelet improved threshold function is proposed. Firstly, this method firstly decomposes the ECG signal by CEEMDAN to obtain a set of intrinsic modal functions (IMFs) from high frequency to low frequency. CEEMDAN decomposition is performed on ECG signal to yield several modal components (IMF). Secondly, the correlation coefficient method is used to perform wavelet denoising with improved threshold on the high frequency IMFs. For the low frequency IMFs, by setting a fixed threshold, the IMFs below the threshold is considered to be the baseline drift signal and removed. Finally, the denoised IMFs and the retained IMFs are reconstructed. The experimental results show that the proposed method is more effective than the empirical mode decomposition (EMD) wavelet denoising, and the global average empirical mode decomposition (EEMD) wavelet denoising method.     

Image denoising based on improved bidimensional empirical mode decomposition thresholding technology

In this paper, a novel image denoising methodology based on improved bidimensional empirical mode decomposition and soft interval thresholding technique is proposed. First, a noise compressed image is constructed. Then, the noise compressed image is decomposed by means of bidimensional empirical mode decomposition (BEMD) into a series of intrinsic mode functions (IMFs), which are separated into signal-dominant IMFs and noise-dominant IMFs using a similarity measure based on ?2-norm and a probability density function, and a soft interval thresholding technique is used adaptively to remove the noise inherent in noise-dominant IMFs. Finally, a denoised image is reconstructed by combining the signal-dominant IMFs and the denoised noise-dominant IMFs. The performance of the proposed denoising method is evaluated by using multiple images with different types of noise, and results from the proposed method are compared with those of other conventional methods in various noisy environments. Simulation results demonstrate that the proposed denoising method outperforms other denoising methods in terms of peak signal-to-noise ratio, mean square error and energy of the first IMF.

     

基于EMD-R/S分析的太赫兹光谱降噪

针对太赫兹时域光谱系统由于延时线的重合抖动、采样抖动等产生的噪声,提出使用经验模态分解-R/S分析方法对太赫兹光谱信号进行降噪.采集太赫兹时域光谱系统的时域信号,根据EMD算法将信号分解成本征模态函数(IMF);使用R/S分析法分别计算各个IMF的Hurst指数.根据设定的阈值判断是否各个IMF是否存在均值回复的情况....     

A novel intelligent denoising method of ecg signals based on wavelet adaptive threshold and mathematical morphology

Due to high-frequency noise and low-frequency noise in ECG signals will interfere with the accurate diagnosis of cardiovascular diseases. With the intrinsic mode function (IMF), which is the main component indicators of high-frequency noise and low-frequency noise, this paper proposes an intelligent denoising method of ECG signals based on wavelet adaptive threshold and mathematical morphology. Firstly, this method performs Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) for signals containing noise, and adopts zero-crossing rate to identify IMFs containing high-frequency noise and low-frequency noise. Secondly, according to the discreteness and randomness of IMF containing high-frequency noise, a wavelet adaptive threshold mathematical model is constructed. In this model, with the signal-to-noise ratio (SNR) improvement as the threshold adjustment parameter, the wavelet threshold is modified by niche genetic algorithm, and the optimal solution is obtained after removing high-frequency noise by wavelet decomposition and reconstruction. The waveform of IMF containing low-frequency noise changes slowly and its amplitude is large and it is difficult to remove low-frequency noise. Therefore, mathematical morphology is used to remove low-frequency noise. Finally, the intelligent denoising method of ECG signals is designed by superimposing denoised IMFs. MIT-BIH experiments show that in the process of removing high-frequency noise and low-frequency noise, compared with other denoising methods, the percent root mean square difference (PRD) and SNR improvement of the method proposed in this paper are improved, and the denoising effect is significant, which can provide expert knowledge and decision-making guidance for related application fields.

     

基于Hilbert-Huang变换的ECG信号降噪方法

提出了一种新的ECG信号降噪方法.该方法以Hilbert-Huang变换(HHT)和阈值降噪方法为核心,将经验模态分解(EMD)分解出的各层固有模态函数(IMF)分为噪声成分起主导作用层和有用信号起主导作用层.对噪声成分起主导作用层采用Donoho等人提出的阈值估计法,对有用信号起主导作用层的噪声水平通过高频层的能量与平均周期的乘积来确定.讨论了噪声成分起主导作用层和有用信号起主导作用层的区分判别问题.该方法克服了阈值估计法对有用信号起主导作用层噪声水平估计的较大偏差.经实验验证表明,该方法能有效地滤除ECG信号检测中的几类主要噪声,且失真很小.     

三维几何信号经验模式分解及其应用

针对三维几何信号非线性、非平稳的特点,提出基于经验模式分解的三维几何信号处理方法。将信号球面参数化,映射到平面,进行均匀规则采样。对平面信号进行限领域的经验模式分解,得到各个内蕴模式图层。从图层信号得到不规则的原始映射信号,逆映射回三维几何模型信号。将该方法用于几何模型的光顺及增强处理,实验结果表明,该方法能够有效处理三维几何信号。     

基于EEMD和二代小波变换的表面肌电信号消噪方法

为了更好地消除混杂在表面肌电信号(sEMG)中的噪声,提出了一种基于总体平均经验模式分解(EEMD)和二代小波变换的sEMG消噪新方法。首先对信号加入白噪声处理后进行经验模态分解(EMD),然后对高频的内蕴模式函数(IMF)分量进行二代小波阈值消噪处理,最后把处理后的高频IMF分量与低频IMF分量进行叠加,重构后的信号即为去噪信号。实验结果表明,该方法融合了二代小波与EEMD的优点,能更好的消除噪声,最大限度的保留有用信号,并具有更高的信噪比。