基于表面肌电信号的腰背动作识别新方法

为提高基于表面肌电信号的人体腰背动作识别率,提出一种基于小波包能量与改进NARX神经网络的分类识别新方法。利用小波包变换对动作部位进行表面肌电信号特征提取,并采用改进NARX神经网络进行分类识别。选取8名实验者分别在扭腰、弯腰、侧弯腰3种动作下进行表面肌电信号数据采集,选择db4小波包函数对信号进行6层分解,得到第6层64个频带的小波包分解系数,代表各个动作信息的特征向量,作为改进NARX神经网络的输入进行分类识别。对照实验组中,改进NARX神经网络的识别率较高,总体识别率达到96.7%。实验结果表明,利用该识别方法对腰部动作进行分类识别,分类准确,且识别率更高。     

基于谐波小波包和支持向量机的声发射源识别

为了解决直升机动部件疲劳损伤类型识别问题,提出了一种基于谐波小波包特征提取和层次支持向量多分类器的声发射源类型识别方法.声发射信号经过4层谐波小波包分解后,提取各个频段的能量特征用于声发射源类型识别,克服了传统小波包分析能量泄露、频带选取不灵活、不同层频率分辨率不同的缺点.首先,利用已知声发射源类型的试验数据训练层次支...     

Joint application of feature extraction based on EMD-AR strategy and multi-class classifier based on LS-SVM in EMG motion classification

This paper presents an effective and efficient combination of feature extraction and multi-class classifier for motion classification by analyzing the surface electromyografic(sEMG) signals. In contrast to the existing methods,considering the non-stationary and nonlinear characteristics of EMG signals,to get the more separable feature set,we introduce the empirical mode decomposition(EMD) to decompose the original EMG signals into several intrinsic mode functions(IMFs) and then compute the coefficients of autoregressive models of each IMF to form the feature set. Based on the least squares support vector machines(LS-SVMs) ,the multi-class classifier is designed and constructed to classify various motions. The results of contrastive experiments showed that the accuracy of motion recognition is improved with the described classification scheme. Furthermore,compared with other classifiers using different features,the excellent performance indicated the potential of the SVM techniques embedding the EMD-AR kernel in motion classification.     

基于小波变换的地震信号去噪方法研究

提高信噪比是地震信号数字处理的主要任务．研究了几种常用的小波去噪方法在地震信号中去除随机噪声的应用,分析了各自的特点,并改进了算法．通过MATLAB的仿真结果表明,小波变换的去噪效果明显优于传统的Fourier变换方法．     

基于分形维前臂动作表面肌电信号的分类

通过分形维对表面肌电信号进行识别分类.在30个健康志愿者做前臂内旋和外旋时,从他们的右前臂肌前群分别采集2类动作表面肌电信号.当原始动作表面肌电信号用小波包变换分解成几个子信号后,采用一种基于模糊自相似性的方法计算原始信号和4个子信号的分形维.结果表明:从频带0～125 Hz的子信号求得的内旋和外旋动作表面肌电信号的分形维有各自的范围;通过该分形维进行Bayes决策时,错误识别率仅2.26%.因此,该分形维适合用来识别内旋和外旋动作表面肌电信号.     

Investigation on the automatic parameters extraction of pulse signals based on wavelet transform

This paper analyses a key problem in the quantification of pulse diagnosis. Due to the subjectivity and fuzziness of pulse diagnosis,quantitative methods are needed. To extract the parameters of pulse signals,the prerequisite is to detect the corners of pulse signals correctly. Up to now,the pulse parameters are mostly acquired by marking the pulse corners manually,which is an obstacle to modernize pulse diagnosis. Therefore,a new automatic parameters extraction approach for pulse signals using wavelet transform is presented. The results testified that the method we proposed is feasible and effective and can detect corners of pulse signals accurately,which can be expected to facilitate the modernization of pulse diagnosis.     

Wavelet packet feature selection for lung sounds based on optimization

In this paper, a wavelet packet feature selection method for lung sounds based on optimization is proposed to obtain the best feature set which maximizes the differences between normal lung sounds and abnormal lung sounds (sounds with wheezes or rales). The proposed method includes two main steps: Firstly, the wavelet packet transform (WPT) is used to extract the original features of lung sounds; then the genetic algorithm (GA) is used to select the best feature set. The obtained optimal feature set is sent to four different classifiers to evaluate the performance of the proposed method. Experimental results show that the feature set obtained by the proposed method provides a higher classification accuracy of 94.6% in comparison with the best wavelet packet basis approach and multi-scale principal component analysis (PCA) approach. Meanwhile, the proposed method has effective generalization performance and can obtain the best feature set without priori knowledge of lung sounds.     

基于小波包分析的滚动轴承声发射信号消噪处理

本文叙述了将声发射检测技术应用于火车货车滚动轴承故障诊断中,通过小波包分析提取滚动轴承声发射信号的能量特征,并通过小波包消噪处理法突出能量特征,以达到较好故障诊断效果。     

Blind separation of speech signals based on wavelet transform and independent component analysis

Speech signals in frequency domain were separated based on discrete wavelet transform （DWT） and inde- pendent component analysis （ICA）. First, mixed speech signals were decomposed into different frequency domains by DWT and the subbands of speech signals were separated using ICA in each wavelet domain; then, the permutation and scaling problems of frequency domain blind source separation （BSS） were solved by utilizing the correlation between adjacent bins in speech signals; at last, source signals were reconstructed from single branches. Experiments were car- ried out with 2 sources and 6 microphones using speech signals at sampling rate of 40 kHz. The microphones were aligned with 2 sources in front of them, on the left and right. The separation of one male and one female speeches lasted 2.5 s. It is proved that the new method is better than single ICA method and the signal to noise ratio is improved by 1 dB approximately.     

基于小波变换的电力滤波器信号调整与分析

本文介绍了小波变换原理,通过对有源电力滤波器的通道信号分解与重构、对谐波进行处理,采用MATLAB工具的函数模块对其通道信号作了仿真分析,结果表明小波变换的多分辨率特性能有效修正仪器通道谐波畸变波及频谱,对进一步实现滤除谐波问题有指导意义。     

Nonlinear multifunctional sensor signal reconstruction based on least squares support vector machines and total least squares algorithm

Least squares support vector machines （LS-SVMs） are modified support vector machines （SVMs） that involve equality constraints and work with a least squares cost function, which simplifies the optimization procedure. In this paper, a novel training algorithm based on total least squares （TLS） for an LS-SVM is presented and applied to multifunctional sensor signal reconstruction. For three different nonlinearities of a multifunctional sensor model, the reconstruction accuracies of input signals are 0.00136%, 0.03184% and 0.504 80%, respectively. The experimental results demonstrate the higher reliability and accuracy of the proposed method for multifunctional sensor signal reconstruction than the original LS-SVM training algorithm, and verify the feasibility and stability of the proposed method.     

基于小波方法的超分辨率图像重建

提出了一种基于小波变换理论的超分辨率重建算法,即利用小波变换得到图像的高频和低频子带,结合非线性外推技术对高频子带进行处理,在增加高频子带信息量的同时进行迭代改进,并采用小波阈值方法进行去噪处理.实验结果表明：该算法能够克服以往插值算法的不足,如高频损失、细节模糊等,能很好地提高图像的峰值信噪比,是图像重建的一种有效方法.     

基于复小波包和turbo码的STBC-MC-CDMA系统及其性能

在分析空时编码技术和多载波CDMA技术原理的基础上,采用turbo编码作为信道编码以及优化的复小波包作为多载波调制,提出一种基于复小波包和turbo码的空时分组编码的MC-CDMA系统,研究了其在瑞利衰落信道下的误码率性能.该系统能充分利用空时分组编码的发送分集和turbo码的良好抗信道衰落能力显著提高系统性能,而且还能利用优化复小波包的优良特性避免通常MC-CDMA系统由于插入循环前缀(CP)所带来频谱效率的降低.仿真结果表明,基于复小波包的空时分组编码的MC-CDMA系统要好于空时分组编码的通常MC-CDMA(STBC-MC-CDMA)系统,略好于采用CP的STBC-MC-CDMA系统;级联turbo码的空时编码技术的应用进一步增强了系统抗衰落信道下的各种干扰能力.     

基于EEMD和LS-SVM的癫痫脑电信号识别

针对癫痫脑电(EEG)信号的非平稳性和非线性,提出一种基于集合经验模式分解(EEMD)提特征并利用最小二乘支持向量机(LS-SVM)的脑电信号分类方法。首先利用EEMD将EEG信号分成多个经验模式分量,得到各阶本征模式分量(IMF),然后提取有效特征,最后用LS-SVM对其进行分类,实验结果表明,该方法对癫痫发作间歇期和发作期EEG的提特征后分类识别正确率达到98%。     

基于小波变换的低比特率图像编码研究

运用小波变换理论,结合人类视觉系统特性,提出一种基于SPIHT算法的低比特率图像压缩的改进算法.实验证明,该算法在平滑图像的压缩中有较好效果,在低比特率压缩下减少了Gibbs现象,边缘模糊现象得到改善.     

基于小波变换和梯度方向的脱机手写藏文字符特征提取方法(英文)

为了提高脱机手写藏文字符的识别效果,提出了一种在小波变换基础上计算局部梯度方向直方图的特征提取方法.首先,对一个脱机手写藏文字符样本图像进行一次Haar小波变换,得到相应的一级近似分量;然后,将这个一级近似分量划分成几个等尺寸的子区域;最后,计算每个等尺寸子区域的局部梯度方向直方图,并将所有子区域的全部局部梯度方向直方图的值作为该字符图片的特征.在最近建立的脱机手写藏文字符样本数据库(THCDB)上的实验结果表明:提出的特征提取方法识别效率较高,且识别效果较好;和细节分量相比,近似分量对提高识别精度具有更大的贡献.     

基于支持向量机的污水状态识别

考虑将支持向量机的思想应用于信息融合,提出基于支持向量机的信息融合的方法,并将这种方法应用于城市污水处理厂的数据处理。同时使用神经网络的信息融合方法与其相比较,实验的结果表明,基于支持向量机的信息融合的方法在对城市污水处理厂的这类数据的模式识别中,有着较好的分类预测能力。     

基于小波变换的跨膜蛋白跨膜螺旋区段预测

1 Introduction In existing genome databases ,about 20 %-30 %ofgenetic products have been esti mated as encodedmembrane proteins[1]. However , about 30 % of allproteins are membrane proteins inliving cells .The TMproteins play ani mportant role especiallyi…     

基于小波变换和径向基神经网络的签名识别

主要研究利用小波变换和径向基神经网络进行签名图像的分类识别.它包括不同签名图像和相似签名图像的分类识别.所提出的方法包括小波域的图像特征提取和利用径向基神经网络的模式分类.采用小波的多分辨分析方法对签名图像进行时频分析特别有效.熵和能量相关特征的概念用于小波域.径向基神经网络具有快速的收敛速度和分类能力.实验仿真证实了利用小波变换和径向基神经网络进行签名图像分类识别的有效性,且成功识别率100%.     

基于混沌映射、小波变换和RSA加密的数字水印算法

本文提出了一种新的公开算法的数字水印算法。该算法首先应用Logistic映射构造了一个原始图像的子图,其次把DWT变换作用在这个子图上得到两个子带LH1和HL1,然后对这两个子带进行RSA加密并把水印嵌入在这两个被加密的子带上,接着解密这两个子带并通过IDWT变换重构子图,最后按构成子图的顺序把每一个8×8像素的小块放回到原图中相应位置,从而得到了一个嵌入了水印信息的图像。实验结果表明,通过该算法嵌入的水印具有较好的鲁棒性、安全性和不可感知性。