[Chapter 15 wavelet based image compression technology deep learning machine learning image processing application matlab.]

The first 15 Chapter Wavelet based image compression technology –matlab Deep learning algorithm to achieve sorting source code analysis

The main function
main.m

clc; clear all; close all;
filename = fullfile(pwd, 'images', 'cameraman.tif');
x = imread(filename);
num = 2;
[cf_vec, dim_vec] = wavedec_process(x, num, 'haar');
th = 10;
y = waverec_process(cf_vec, dim_vec, 'haar', th);
output_img(x, y, filename, th, 'png');
p = PSNR(x,y);
fprintf('\n Image before and after compression PSNR The value is %.2f\n', p);

Case effect display ：



Function Directory ：

Output subfunction ：

function output_img(x, y, filename, th, ext)
if nargin < 5
    ext = 'png';
end
[~, file, ~] = fileparts(filename);
foldername = fullfile(pwd, 'output');
if ~exist(foldername, 'dir')
    mkdir(foldername);
end
file1 = fullfile(foldername, sprintf('%s_origin.%s', file, ext));
file2 = fullfile(foldername, sprintf('%s_wave_%.1f.%s', file, th, ext));
imwrite(x, file1);
imwrite(y, file2);
info1 = imfinfo(file1);
info2 = imfinfo(file2);
fprintf('\n The storage space required for the image before compression is %.2fbytes', info1.FileSize);
fprintf('\n The storage space required for the compressed image is %.2fbytes', info2.FileSize);
fprintf('\n The file size ratio is %.2f', info1.FileSize/info2.FileSize);

Drawing subfunction 1

function plot_wave_coef(cf_vec)
dn = 3;
num = (length(cf_vec)-1)/dn;
figure;
subplot(num+1, 3, 2);
yt = im2uint8(mat2gray(cf_vec{
    1}));
imshow(yt, []); 
title(sprintf(' Approximate component A_{%d}', num));
info = {
    ' Vertical details V', ' Level of detail H', ' Diagonal details D'};
ps = 2;   
for i = 1 : num
    for j = 1 : dn
        yt = im2uint8(mat2gray(cf_vec{
    ps}));
        subplot(num+1, dn, ps+2);
        imshow(yt, []);
        title(sprintf('%s_{%d}', info{
    j}, num-i+1));
        ps = ps+1;
    end
end

Drawing subfunction 2

function plot_wave_coef_join(cf_vec,dim_vec)
dn = 3;
num = (length(cf_vec)-1)/dn;
tmpa = wkeep(cf_vec{
    1}, dim_vec(1, :), 'c');
tmpa = im2uint8(mat2gray(tmpa));
tmpa(1, :) = 255; tmpa(end, :) = 255;
tmpa(:, 1) = 255; tmpa(:, end) = 255;
for j = 1:num
    tmpv = wkeep(cf_vec{
    (j-1)*dn+2}, dim_vec(j, :), 'c');
    tmph = wkeep(cf_vec{
    (j-1)*dn+3}, dim_vec(j, :), 'c');
    tmpd = wkeep(cf_vec{
    (j-1)*dn+4}, dim_vec(j, :), 'c');
    tmpv = im2uint8(mat2gray(tmpv));
    tmph = im2uint8(mat2gray(tmph));
    tmpd = im2uint8(mat2gray(tmpd));
    tmpv(1, :) = 255; tmpv(end, :) = 255;
    tmpv(:, 1) = 255; tmpv(:, end) = 255;
    tmph(1, :) = 255; tmph(end, :) = 255;
    tmph(:, 1) = 255; tmph(:, end) = 255;
    tmpd(1, :) = 255; tmpd(end, :) = 255;
    tmpd(:, 1) = 255; tmpd(:, end) = 255;    
    tmp = [tmpa,tmpv;tmph,tmpd];
    stc = size(tmp);
    if stc >= dim_vec(j+1, :)
        tmpa = tmp(1:dim_vec(j+1, 1), 1:dim_vec(j+1,2));
    else
        tmp = tmp([1:end-1, end-2:end-1], [1:end-1, end-2:end-1]);
        tmpa = tmp(1:dim_vec(j+1, 1), 1:dim_vec(j+1,2));
    end
    tmpa = im2uint8(mat2gray(tmpa));
    tmpa(1, :) = 255; tmpa(end, :) = 255;
    tmpa(:, 1) = 255; tmpa(:, end) = 255;
end
figure;
imshow(tmpa, []);
title(' Tower graph of wavelet coefficients ');

PSNR
If not used in inverse discrete cosine transform 8 As a coefficient , But use 4 As a coefficient , The image quality will deteriorate . To calculate the peak signal-to-noise ratio of the input image and the image after inverse discrete cosine transform ！ Then the bit rate of inverse discrete cosine transform is obtained
Peak signal to noise ratio （Peak Signal to Noise Ratio） Abbreviation for PSNR, The ratio of the maximum possible power of a signal to the destructive noise power that affects its representation accuracy , It can display the degree of image quality loss .

Tips ： The content is organized from ：

https://github.com/gzr2017/ImageProcessing100Wen

The higher the peak signal-to-noise ratio , Indicates that the smaller the image quality loss . The peak signal-to-noise ratio is defined by the following formula .MAX Represents the maximum value of the image point color . If the range is [0,255] Words , that MAX The value is 255. Reference source

PSNR Processing function

function S=PSNR(s,t)
[m, n, ~]=size(s);
s = im2uint8(mat2gray(s));
t = im2uint8(mat2gray(t));
s = double(s);
t = double(t);
sd = 0;
mi = m*n*max(max(s.^2));
for u = 1:m
    for v = 1:n
        sd = sd+(s(u,v)-t(u,v))^2;
    end
end
if sd == 0
    sd = 1;
end
S = mi/sd;
S = 10*log10(S);

Partial reference ：https://blog.csdn.net/weixin_29732003/article/details/122569893
SNR Is the reciprocal of the coefficient of variation , Same as the coefficient of variation , Eliminate units and （ or ） The influence of different means on the comparison of variation degree of two or more data , At the same time, there may be a corresponding relationship between the numerical value and the visual image quality ：
SNR The higher the , The less discrete the pixels are , image quality （ Probably ） The better .

Code source of this article Arrangement –> Portal