基于图像小波变换的多尺度自适应双边滤波matlab仿真

1.前言

图像处理领域中，噪声抑制与边缘保持的矛盾长期存在。传统方法如高斯滤波通过空间邻域加权实现平滑，但会导致边缘模糊；双边滤波引入像素值相似性约束，虽能保边却难以处理复杂纹理。基于小波变换的多尺度自适应双边滤波（MSABF）通过融合多尺度分析、自适应参数调整与非线性扩散技术，实现了噪声与边缘的智能区分。

2.算法运行效果图预览

(完整程序运行后无水印)

3.算法运行软件版本

Matlab2024b(推荐)或者matlab2022a

4.部分核心程序

（完整版代码包含中文注释和操作步骤视频）

复制代码

for i = 1:12
    i
    Image_Name_In  = ['../D_THZ图像产生程序/THZ_Images/' ,num2str(i),'out.png'];
    Image_Name_Out = ['Result/',num2str(i),'out.png'];

    I0             = imread(Image_Name_In);
    if(size(I0,3)==3)
        I0= rgb2gray(I0);
    end
    I1    = double(I0)/255;

    Iret0 = bfilter2(I1);
    Iret1 = uint8(255*Iret0);

    Iret2 = ImageEnhanceByMUSICAandWAVELET(Iret1,Image_Name_Out);
    
    %计算标准图像和处理后的图像的PSRN值
    STAND_IMAGE  = ['../D_THZ图像产生程序/image/',num2str(i),'out.png'];
    STAND_IMAGE0 = imread(STAND_IMAGE);
    STAND_IMAGE0 = 255*im2bw(STAND_IMAGE0,graythresh(STAND_IMAGE0));
    if(size(STAND_IMAGE0,3)==3)
        STAND_IMAGE0 = rgb2gray(STAND_IMAGE0);
    end
    PSNR(i) = func_PSNR(uint8(STAND_IMAGE0),uint8(Iret2));
    
    figure(i);
    subplot(131);
    imshow(uint8(I0));
    title('处理前');
    subplot(132);
    imshow(uint8(Iret1));
    title('自适应BF');
    subplot(133);
    imshow(uint8(Iret2));
    title('小波增强');    
    pause(1);
end   
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
figure;
plot(PSNR,'b-o');
xlabel('图像编号');
ylabel('PSNR');
axis([0,12,15,40]);
save PSNR.mat PSNR
25_009m