Matlab learning 7- linear smoothing filtering of image processing

Tips ： When the article is finished , Directories can be generated automatically , How to generate it, please refer to the help document on the right

---

Preface

Spatial filtering enhancement

• Convolution principle
  • Multidimensional continuous convolution
• Linear smoothing filtering
  • Domain average method 、 Selective averaging 、Wiener wave filtering
• Nonlinear smoothing filtering
  • median filtering
• Linear sharpening filter
  • Laplacian operator
• Nonlinear sharpening filter
  • Prewitt operator
  • Sobel operator
  • Log operator

Matlab Study 7- Linear smoothing filtering for image processing
Domain average method 、 Selective averaging 、Wiener wave filtering

---

One 、 Select the average method for filtering and denoising

effect

Code

%  Select the average method for filtering and denoising 
img =imread("img/test.bmp");
subplot(2,3,1);imshow(img),xlabel(" original image ");

%  Mean filtering 
K1=filter2(fspecial("average",3),img);
subplot(2,3,2);imshow(uint8(K1)),xlabel(" Original picture 3*3 Convolution and mean filtering ");

% Convolution sum   Or template 
mask=[  0 0 0
        0 1 1
        0 1 1];
mask=1/4*mask;
K2=filter2(mask,img);
subplot(2,3,3);imshow(uint8(K2)),xlabel(" The original image selects average convolution and average filtering ");

ju=imnoise(img,"salt & pepper",0.04);
subplot(2,3,4);imshow(ju),xlabel(" Noise image ");

%  Mean filtering 
K3=filter2(fspecial("average",3),ju);
subplot(2,3,5);imshow(uint8(K3)),xlabel(" Noise image 3*3 Convolution and mean filtering ");
% Convolution sum   Or template 
mask=[  0 0 0
        0 1 1
        0 1 1];
mask=1/4*mask;
K4=filter2(mask,ju);
subplot(2,3,6);imshow(uint8(K4)),xlabel(" Average convolution and mean filtering are selected for noisy images ");

Two 、 Domain average denoising

effect

Code

% Domain average denoising 
img =imread("img/eight.tif");
subplot(2,3,1);imshow(img),xlabel(" original image ");

ju=imnoise(img,"salt & pepper",0.004);
subplot(2,3,2);imshow(ju),xlabel(" Noise image ");
%  Mean filtering 
K1=filter2(fspecial("average",3),ju);
subplot(2,3,3);imshow(uint8(K1)),xlabel("3*3 Convolution and mean filtering ");

K2=filter2(fspecial("average",5),ju);
subplot(2,3,4);imshow(uint8(K2)),xlabel("5*5 Convolution and mean filtering ");

% Convolution sum   Or template 
mask4=[ 0 1 0
        1 0 1
        0 1 0];
mask4=1/4*mask4;
K3=filter2(mask4,ju);
subplot(2,3,5);imshow(uint8(K3)),xlabel("4 Neighborhood convolution and mean filtering ");


mask8=[ 1 1 1
        1 0 1
        1 1 1];
mask8=1/8*mask8;
K4=filter2(mask8,ju);
subplot(2,3,6);imshow(uint8(K4)),xlabel("8 Neighborhood convolution and mean filtering ");

3、 ... and 、Wiener wave filtering

effect

Code

% Wiener wave filtering 
img =imread("img/eight.tif");
subplot(2,3,1);imshow(img),xlabel(" original image ");
ju=imnoise(img,"salt & pepper",0.04);
subplot(2,3,2);imshow(ju),xlabel(" Noise image ");

K1=wiener2(ju,[3,3]);
subplot(2,3,3);imshow(K1),xlabel(" Noise image 3*3Wiener wave filtering ");
K2=wiener2(ju,[5,5]);
subplot(2,3,4);imshow(K2),xlabel(" Noise image 5*5Wiener wave filtering ");
K3=wiener2(ju,[7,7]);
subplot(2,3,5);imshow(K3),xlabel(" Noise image 7*7Wiener wave filtering ");
K4=wiener2(ju,[9,9]);
subplot(2,3,6);imshow(K4),xlabel(" Noise image 9*9Wiener wave filtering ");

Four 、 Linear smoothing filtering

Code

% Linear smoothing filtering 
Fxy=[0 20 40 70
    80 100 120 150
    160 180 200 230];
uint8Fxy=uint8(Fxy);
subplot(2,2,1),imshow(uint8Fxy),xlabel(" original image ");
subplot(2,2,2),imhist(uint8Fxy),axis([0,255,0,1]),xlabel(" Histogram of the original image ","position",[120,-0.23]);

% Smooth operation , Fuzzy action , Denoise 
Gxy=filter2(fspecial("average",3),uint8Fxy);
uint8Gxy=uint8(Gxy);
subplot(2,2,3),imshow(uint8Gxy),xlabel(" Mean filtered image ");
subplot(2,2,4),imhist(uint8Gxy),axis([0,255,0,1]),xlabel(" Histogram of filtered image ","position",[120,-0.23]);

---

Click to get the source code
https://gitee.com/CYFreud/matlab/tree/master/image_processing/demo7_220418