Convert 4096 (8 X 8) image blocks into an image
Krishna Prasad Satamraju
clc;
clear all;
close all;
start_time = cputime; %Starting Time of Execution
I = imread('lena.jpg');
[r, c]=size(I);
bs=8; % Block Size (8x8)
nob=(r/bs)*(c/bs); % Total number of 8x8 Blocks
% Dividing the Cover image into 8x8 Blocks
for i=1:(r/bs)
for j=1:(c/bs)
dctimg(:,:,kk+j)=dct2(I((bs*(i-1)+1:bs*(i-1)+bs),(bs*(j-1)+1:bs*(j-1)+bs)));
end
kk=kk+(r/bs);
end
size(dctimg) = 8 8 4096.
How can I get size(dctimg) = 512 X 512;
Please Help. thanks in advance.
Sean de
doc reshape
but why not just use blkproc to do the whole thing for you?
doc blkproc
or even cell2mat->cellfun->mat2cell
?
ImageAnalyst
> size(dctimg) = 8 8 4096.
> How can I get size(dctimg) = 512 X 512;
> Please Help. thanks in advance.
I don't know. How did you *want* to turn your 3D image into a 2D
image? I have no idea. Maybe summing, maybe reshaping, maybe just
building it up differently in the first place, I have no idea. I
don't even know why you created this 3D image in the first place
instead of doing what Sean suggested. Are you SURE you really need a
3D (8x8x4096) or 2D (512x512) image at all???
Jan Simon
Let me add my wild guessing also:
> > size(dctimg) = 8 8 4096.
> > How can I get size(dctimg) = 512 X 512;
% D = [8 x 8 x 4096]
D = reshape(D, 8, 8, 64, 64);
E = permute(D, [1, 3, 2, 4]);
F = reshape(D, 512, 512);
Perhaps, or perhaps not. Jan
Krishna Prasad Satamraju
My aim is to
1. divide image(512 X 512 lena.jpg) into blocks of 8X8,
2. then perform DCT on each block.
3. In the DCT, replace the midfrequency coefficients (we get 64 coefficients out of which I pick 9) with watermark image. (my watermark image is divided into 3X3 block)
4. After replacing, perform IDCT to get the watermarked image. If I use blkproc I get dct for the entire image and cannot access individual 8 X 8 blocks
That's why I followed the above method, where I can access each block individually
The dctimg is an array of 8X8 dct coefficients.
size(dctimg) = 8 8 4096
Now I have to combine the 4096 blocks of image into a single image and calculate inverse DCT on it.
Please suggest a solution thank you
ramya s
My aim is to
1. divide image(512 X 512 lena.jpg) into blocks of 8X8,
2. then perform DCT on each block and find quantized DCT values..
3. modify the non zero DCT quantized values coefficients (we get only 6-9 coeff out of 64 coefficients) with secret message.
4. After modifying, perform IDCT to get the stego image.
If I use blkproc I get dct for the entire image and cannot access individual 8 X 8 blocks
please help me how to proceed with the 8x8 block.
> > > clc;
> > > clear all;
> > > close all;
> > >
> > > start_time = cputime; %Starting Time of Execution
> > >
> > > I = imread('lena.jpg');
> > >
> > > [r, c]=size(I);
> > > bs=8; % Block Size (8x8)
> > > nob=(r/bs)*(c/bs); % Total number of 8x8 Blocks
> > >
> > > % Dividing the Cover image into 8x8 Blocks
> > > for i=1:(r/bs)
> > > for j=1:(c/bs)
> > > dctimg(:,:,kk+j)=dct2(I((bs*(i-1)+1:bs*(i-1)+bs),(bs*(j-1)+1:bs*(j-1)+bs)));
> > > end
> > > kk=kk+(r/bs);
> > > end
> > > size(dctimg) = 8 8 4096.
> > > How can I get size(dctimg) = 512 X 512;
> My aim is to
eric paulraj
HI, CAN I GET THE SOLUTION FOR DIVIDING THE 512X512 IMAGE INTO 8X8 BLOCK
