%% init

dim = 256;

imgnum = 20;

wmlength = 1024;

dimblk = 32; % block dimension

wmnum = 1000;

load imgs.mat

% imgs = zeros(dim, dim, imgnum);

% for i = 1:imgnum

% imgFile = ['image',num2str(i),'.bmp'];

% % imgs(:,:,i) = imresize(imread(['images/',imgFile]), [dim dim]);

% imgs(:,:,i) = im2double(imresize(imread(['images/',imgFile]), [256 256]));

% end

load DCTblks.mat

% DCTblks = compDCTblks(imgs, imgnum);

load pnseqs.mat

% shift = round(rand() * wmnum);

% pnseqs = genPN(wmlength, shift);

% idx = randperm(4096);

% 1:2730, 2731:end (2/3 for parameter estimation, 1/3 for model testing)

%% main process

% 20 images, 1000 watermarks, 9 embed positions, 4 groups of data, 50 thresholds.

% At every execution we calculate the corresponding test statistics

% with and without the watermark.

D = dctmtx(8);

img2dct = @(block_struct) D * block_struct.data * D';

dct2img = @(block_struct) D' * block_struct.data * D;

embedScheme = 'DFT';

fftobj = vision.FFT;

ifftobj = vision.IFFT;

attackCategory = 'jpeg';

imgATT0 = attack(imgs, attackCategory);

coef0 = zeros(size(imgATT0));

for i = 1:imgnum

coef0(:,:,i) = abs(fftshift(step(fftobj, squeeze(imgATT0(:,:,i)))));

end

% params = getParams(ATTDCTblks0, lktbl);

% params: sitenum X imgnum X modelnum X 2

result = zeros(wmnum, imgnum, 2, 2);

masksize = wmlength;

[mask, mask_symmetry] = compMaskDFT(dim,dim,masksize);

for j = 1:wmnum

wm = pnseqs(1:wmlength, j);

disp([' watermark: ', num2str(j)]);

% tic;

embedDFT;

imgATT1 = attack(imgWM, attackCategory);

detectDFT;

% toc;

% return;

end

filename = ['result_', embedScheme, attackCategory, '2.mat'];

save(filename, 'result');