阿尔法均值滤波汇编.docVIP

图像复原的目的是从观测到的退化图像重建原始图像，它是图像处理、模式识别、机器视觉等的基础，在天文学、遥感成像、医疗图像等领域获得了重要应用。 代码： I=imread(g.png); [m,n]=size(I); P=imnoise(I,salt pepper,0.04); Q=imnoise(I,gaussian,0,0.02); R=imnoise(P,gaussian,0,0.02); %加椒盐噪声的三种情况 A=I;%d=0 d=0; for i=2:m-1 for j=2:n-1 img_turn = P(i-1:i+1,j-1:j+1); img_turn = sort(img_turn(:)); img_decrese = img_turn(floor(d/2)+1:9-floor(d/2)); A(i,j) = sum(img_decrese(:))/(9-d); end end B=I;%d=2 d=2; for i=2:m-1 for j=2:n-1 img_turn = P(i-1:i+1,j-1:j+1); img_turn = sort(img_turn(:)); img_decrese = img_turn(floor(d/2)+1:9-floor(d/2)); B(i,j) = sum(img_decrese(:))/(9-d); end end C=I;%d=4 d=4; for i=2:m-1 for j=2:n-1 img_turn = P(i-1:i+1,j-1:j+1); img_turn = sort(img_turn(:)); img_decrese = img_turn(floor(d/2)+1:9-floor(d/2)); C(i,j) = sum(img_decrese(:))/(9-d); end end %加高斯噪声的三种情况 D=I;%d=0 d=0; for i=2:m-1 for j=2:n-1 img_turn = Q(i-1:i+1,j-1:j+1); img_turn = sort(img_turn(:)); img_decrese = img_turn(floor(d/2)+1:9-floor(d/2)); D(i,j) = sum(img_decrese(:))/(9-d); end end E=I;%d=2 d=2; for i=2:m-1 for j=2:n-1 img_turn = Q(i-1:i+1,j-1:j+1); img_turn = sort(img_turn(:)); img_decrese = img_turn(floor(d/2)+1:9-floor(d/2)); E(i,j) = sum(img_decrese(:))/(9-d); end end F=I;%d=4 d=4; for i=2:m-1 for j=2:n-1 img_turn = Q(i-1:i+1,j-1:j+1); img_turn = sort(img_turn(:)); img_decrese = img_turn(floor(d/2)+1:9-floor(d/2)); F(i,j) = sum(img_decrese(:))/(9-d); end end %加混合噪声的三种情况 G=I;%d=0 d=0; for i=2:m-1 for j=2:n-1 img_turn = R(i-1:i+1,j-1:j+1); img_turn = sort(img_turn(:)); img_decrese = img_turn(floor(d/2)+1:9-floor(d/2)); G(i,j) = sum(img_decrese(:))/(9-d); end end H=I;%d=2 d=2; for i=2:m-1 for j=2:n-1 img_turn = R(i-1:i+1,j-1:j+1); img_turn = sort(img_turn(:));