www.gusucode.com > 数字水印算法实现(matlab程序包),包含有DCT水印嵌入算法源码程序 > watermark综述+代码/cor_recover.m
%Name: Chris Shoemaker %Course: EER-280 - Digital Watermarking %Project: Threshold-Based Correlation using block processing in the spatial domain % Watermark Recovery clear all; % save start time start_time=cputime; blocksize=16; % set the size of the block in cover to be used for each bit in watermark % read in the watermarked object file_name='cor_watermarked.bmp'; watermarked_image=double(imread(file_name)); % determine size of watermarked image Mw=size(watermarked_image,1); %Height Nw=size(watermarked_image,2); %Width % determine maximum possible message size in object max_message=Mw*Nw/(blocksize^2); % read in original watermark file_name='_copyright.bmp'; orig_watermark=double(imread(file_name)); % determine size of original watermark Mo=size(orig_watermark,1); %Height No=size(orig_watermark,2); %Width % read in key for PN generator file_name='_key.bmp'; key=double(imread(file_name))./256; % reset MATLAB's PN generator to state "key" rand('state',key); % generate the watermark equal to the size of one block pn_sequence=round(2*(rand(blocksize,blocksize)-0.5)); % pad message out to maximum message size with ones message_vector=ones(No*Mo,1); % process the image in blocks % for each block determine it's correlation with base pn sequence x=1; y=1; for (kk = 1:length(message_vector)) % sets correlation to 1 when patterns are identical to avoid /0 errors % otherwise calcluate correlation if (watermarked_image(y:y+blocksize-1,x:x+blocksize-1) == pn_sequence) correlation(kk)=1; else correlation(kk)=corr2(watermarked_image(y:y+blocksize-1,x:x+blocksize-1),pn_sequence); end % move on to next block. At and of row move to next row if (x+blocksize) >= Nw x=1; y=y+blocksize; else x=x+blocksize; end end % if correlation exceeds average correlation for kk = 1:length(correlation) if (correlation(kk) > mean(correlation(1:Mo*No))) message_vector(kk)=0; end end % reshape the message message=reshape(message_vector(1:Mo*No),Mo,No); % display processing time elapsed_time=cputime-start_time, % display the recovered message figure(2) imshow(message,[]) title('Recovered Message')