How can I reconstruct a color image to get its original version using wavelet?
1 次查看(过去 30 天)
显示 更早的评论
Wavelet is considered to be completely invertible. That is if an image is decomposed using wavelet. then the image can be get back by reconstructing it.
I am trying with this code for an color image. I perform selective reconstruction and then perform addition of those reconstructed details. But failing to recover completely the original image .
Please provide me with some suggestions or some code in matlab.
code :
I=imread('mouse.jpg'); %I=rgb2gray(I); %I=randn(256,256);
[Lo_D,Hi_D,Lo_R,Hi_R]=wfilters('db8');
[C,S]=wavedec2(I,2,Lo_D,Hi_D);
LF1=wrcoef2('a',C,S,'db8',1);
LF2=wrcoef2('a',C,S,'db8',2);
HHF1=wrcoef2('h',C,S,'db8',1);
VHF1=wrcoef2('v',C,S,'db8',1);
DHF1=wrcoef2('d',C,S,'db8',1);
HHF2=wrcoef2('h',C,S,'db8',2);
VHF2=wrcoef2('v',C,S,'db8',2);
DHF2=wrcoef2('d',C,S,'db8',2);
LF1=wcodemat(LF1,192); HHF1=wcodemat(HHF1,192); VHF1=wcodemat(VHF1,192); DHF1=wcodemat(DHF1,192);
LF2=wcodemat(LF2,192); HHF2=wcodemat(HHF2,192); VHF2=wcodemat(VHF2,192); DHF2=wcodemat(DHF2,192);
figure(1); title('low pass') subplot(2,4,1); imshow(uint8(LF1)); title('A1'); %axis image; % Display equal on x and y axes %axis off;
subplot(2,4,2); imshow(uint8(HHF1)); title('H1');
subplot(2,4,3); imshow(uint8(VHF1)); title('V1');
subplot(2,4,4); imshow(uint8(DHF1)); title('D1');
subplot(2,4,5); imshow(uint8(LF2)); title('A2');
subplot(2,4,6); imshow(uint8(HHF2)); title('H2');
subplot(2,4,7); imshow(uint8(VHF2)); title('V2');
subplot(2,4,8); imshow(uint8(DHF2)); title('D2');
%summatiomn L1=uint8(LF1); H1=uint8(HHF1); V1=uint8(VHF1); D1=uint8(DHF1);
result=L1+H1+V1+D1;
figure(2),imshow(result),title('Recovered image from L1,H1,V1,D1');
imwrite(result,'recover1.jpg','jpg');
%summatiomn L2=uint8(LF2); H2=uint8(HHF2); V2=uint8(VHF2); D2=uint8(DHF2);
result2=L2+H2+V2+D2;
figure(3),imshow(result2),title('Recovered image from L2,H2,V2,D2');
figure(4),imshow(I),title('Original Image');
回答(0 个)
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 AI for Signals and Images 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!