PCA.zip_PCA 遥感_PCA融合_pca图像融合_timesld_图像融合算法

%PCA clear up=imread('high.jpg'); %读图像 low=imread('low.jpg'); figure(1) imshow(up); %读RGB数值 title('PCA-RGB表示的高分辨率图像'); figure(2) imshow(low); title('PCA-RGB表示的低分辨率图像'); up=double(up); low_R=double(low(:,:,1)); low_G=double(low(:,:,2)); low_B=double(low(:,:,3)); [M,N,color]=size(up); low_Mx=[0 0 0]'; for i = 1 : M for j = 1 : N low_S = [low_R(i,j),low_G(i,j),low_B(i,j)]'; low_Mx = low_Mx + low_S; end end low_Mx = low_Mx / (M*N);% 计算低分辨率图片的三维列向量的平均值 low_Cx=0; for i = 1 : M for j = 1 : N low_S = [low_R(i,j),low_G(i,j),low_B(i,j)]'; low_Cx = low_Cx + low_S*low_S'; end end low_Cx = low_Cx / (M * N)- low_Mx*low_Mx'; % 运用公式cov(x,y)=E(xy)-ExEy 计算协方差矩陈 [low_A,low_latent] = eigs(low_Cx);% 协方差矩陈的特征向量组成的矩陈----PCA变换的系数矩陈,特征值. for i = 1 : M for j = 1 : N low_X = [low_R(i,j),low_G(i,j),low_B(i,j)]'; low_Y = low_A'*low_X; low_Y=low_Y'; low_R(i,j) = low_Y(1); % 低分辨率图片的第1主分量 low_G(i,j) = low_Y(2); % 低分辨率图片的第2主分量 low_B(i,j) = low_Y(3); % 低分辨率图片的第3主分量 end end im=uint8(low_R); %匹配一个图像的直方图 y=imhist(im); y=y/sum(y); %归一化，使函数符合概率分布的sum(y)==1这样一个规律 plot(y); %待匹配的直方图 G=[]; %函数的累积直方图 for i=1:256 G=[G sum(y(1:i))]; end img=imread('high.jpg'); [m n]=size(img); hist=imhist(img); %待处理图像的直方图 p=hist/(m*n); s=[]; %待处理图像的累积直方图 for i=1:256 s=[s sum(p(1:i))]; end for i=1:256 tmp{i}=G-s(i); tmp{i}=abs(tmp{i}); %因为要找距离最近的点，所以取绝对值 [a index(i)]=min(tmp{i}); %找到两个累积直方图距离最近的点 end imgn=zeros(m,n); for i=1:m for j=1:n imgn(i,j)=index(img(i,j)+1)-1; %由原图的灰度通过索引映射到新的灰度 end end imgn=uint8(imgn); imgn=double(imgn); for i = 1 : M for j = 1 : N low_Y = [imgn(i,j),low_G(i,j),low_B(i,j)]';%将低分辨率的全色图像的第一主成分替换成高分辨率的图片 low_X = low_A*low_Y; low_X=low_X'; r(i,j) = low_X(1); g(i,j) = low_X(2); b(i,j) = low_X(3); end end RGB(:,:,1)=r; RGB(:,:,2)=g; RGB(:,:,3)=b; figure(3) imshow(uint8(round(RGB))) title('PCA-RGB表示的转化图像');