资源简介
基于多算法融合的小波降噪方法,可以实现一维信号的有效去噪,另外附有盲源分离代码!
代码片段和文件信息
%FastICA算法代码
%对数据进行标准化
close all
load leleccum;
s = leleccum(1:1000);
%load noisbloc; s = noisbloc;
figure(1);
plot(s);
title(‘原始信号‘);
%grid on;
xlabel(‘样本序号 n‘);
ylabel(‘幅值 A‘);
N = length(s);
k = 5;
[cl]=wavedec(sk‘db6‘);
%Y = appcoef(cl‘db6‘k);
%Y1 = appcoef(cl‘db6‘k);
%Y2 = appcoef(cl‘db6‘k);
%Z = appcoef(cl‘db6‘k);
%for i = 1:k
%dx = detcoef(cli);
N = length(s);
%N = 1000;
j = randint(11[1k]);
%{
sigma=wnoisest(cl6);%估计一维小波的系数的标准偏差
display(sigma);
y=2*log(N)*sigma
thr=sqrt(y);
%}
%A = median(abs(dx))/0.6745;
%{
A = median(abs(dx))/0.6745;
T = sqrt(2*log10(N))/log10(j+1);
thr = A*T;
%}
%thr = thselect(dx‘heursure‘);%SURE阈值选取
%{
sigma = wnoisest(cl6);
alpha = 2;
thr = wbmpen(clsigmaalpha);%全局阈值
%}
%{
Q = sqrt(sum(abs(dx).^2)./N.^2);
y = max(Q.^2-A.^2eps);
C=sqrt(y);
thr=A.^2./C;
%}
S = ones(1k);
first = cumsum(l)+1;
first = first(end-2:-1:1);
ld = l(end-1:-1:2);
last = first+ld-1;
cxd = c;
lxd = l;
for i = 1:k
flk = first(i):last(i);
if S(i) < sqrt(eps) * max(c(flk))
%A = median(abs(c(flk)))/0.6745;
% thr = thselect(c(flk)‘heursure‘);
thr = 0;
%xd = c(flk);
%T = A*sqrt(2*log(N));
%thr = T/log(j+1);
else
thr = thselect(c(flk)/S(i)‘heursure‘);%SURE阈值选取
end
%{
A = median(abs(xd))/0.6745;
T = A*sqrt(2*log(N));
thr = T/log(j+1);
%}
%A = median(abs(c))/0.6745;
%thr = A*sqrt(2*log(N));
%{
Q = sqrt(sum(abs(c).^2)./N.^2);
y = max(Q.^2-A.^2eps);
C=sqrt(y);
thr=A.^2./C;
%}
thr = thr * S(i);
%u = sqrt((abs(c(flk))-thr)./(abs(c(flk))+thr));
%a = thr./(norm(c(flk))*norm(u));
a = 0.6;
%end
xd = c(flk);
%}
%A = (abs(dx)-thr)/N;
%var = thr./exp(A);
tmp = abs(xd)-a*thr;
tmp = (tmp+abs(tmp))/2;
value = (1-a)*sign(xd).*tmp;
%}
B = 2*xd./thr;
def = 2*thr./(1+exp(B));
tmp1 = xd-thr+def;
value1 = a.*(tmp1);
cxd(flk) = value+value1;%新阈值处理法
%}
%Y = [Y D];
%{
tmp = abs
- 上一篇:Matlab求解Var代码
- 下一篇:形态滤波的matlab代码
相关资源
- Wavelet power load forecasting 基于小波神经
- 改进型的小波包与PNN的联合故障诊断
- Wavelet_OMP
- Wavelet_denoising 小波变换的MATLAB源程序
-
thresholddenoisingba
sedonwavelet 小波阈值 - image-fusion-and-detection-on-curvelet-wavelet
- waveletdenoising 小波变换应用
- wavelet 小波变换用于脑电信号处理
- wavelet-matlab 基于小波变换的图像融合
- CS-OMP-wavelet-reconstruction
- Stationary-wavelet-transform
- EEG_WAVELET 用小波方法把脑电信号进行
- Inpainting--on-Wavelet- 图像修复是图像处
- Wavelet-Tools 美国Rice大学提供的matlab 小
- Wavelet-network-timeseries 使用小波神经网
- Wavelet_IRLS 压缩感知CS——采用小波变
- Wavelet_EntropyinformationLZC 脑电处理中
- advanced-harmonic-wavelet-packet 自己编的改
- wavelet 用于求取小波包变换后小波系数
- Wavelet-moment-to-compile
- wavelet-threshold
- Wavelet-transform 本程序利用小波变换的
- image-denoising-using-wavelet 该文件夹包括
- synsq_toolbox_v1 同步压缩小波变换(sy
- fusion_and_evaluate
- Wavelet_ROMP 压缩感知CS——采用小波变
- Ga_Wavelet_Net
- WaveletEnergySpectrum
- wavelet 用小波变换进行图像增强
- matlab; GUI; wavelet
评论
共有 条评论