1. Algorithm part simulation preview

        In order to adapt to the non-stationary changes in the spatial domain of the image , A classification hidden Markov tree based on wavelet domain is proposed (CHMT) Image denoising method of model . In the model , The wavelet coefficients of each subband of the image at each scale are divided into C Group to highlight the non-stationary changes in its airspace Characteristics of , In this way, the original wavelet quadtree is divided into C Trees have different HMT Parameter wavelet quadtree , After reasonable initialization and expectation maximization (EM) Algorithm training parameters , Inverse transformation recovery . experiment It turns out that , Compared with existing methods , This method does not increase the amount of calculation , The quality of the restored image is significantly improved (PSNR).

clc;
clear;
close all;
warning off;
addpath(genpath(pwd));


load lena;
 
sigma = 0.1; %noise standard deviation
hh = daubcqf(4); %wavelet filter

x = lena + sigma*randn(size(lena));
disp(['PSNR of noisy image is ' num2str(psnr(lena,x)) 'dB']);
y=hdenoise(x,hh);
disp(['PSNR of denoised image is ' num2str(psnr(lena,y)) 'dB']);
figure(1);
image(x*255+1);
colormap(gray(256));
axis square;
title('Noisy image');
figure(2);
image(y*255+1);
colormap(gray(256));
axis square;
title('Denoised image');

2. Source code acquisition method

Get the way 1：

Click the download link ：

【 Provider operation video 】 Research on image denoising method based on Hidden Markov tree model in wavelet domain matlab Implementation code

Access method 2：

Blog resource item , Search for resources with the same name as blog .

Access method 3：

If the download link fails , Blogger wechat contact .