基于RPCA异常值检测代码.zip

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % This is a sample code for testing the algorithm in our paper % ''Fast Alternating Linearization Methods for % Minimizing the Sum of Two Convex Functions'', Donald Goldfarb, % Shiqian Ma and Katya Scheinberg, Tech. Report, Columbia University, % 2009 - 2010. Preprint available at: % http://arxiv.org/pdf/0912.4571v2.pdf % % Author: Shiqian Ma % Date : Apr. 20, 2010 % IEOR, Columbia University, Copyright (2010) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % get data tic; dataformat = 'surveillance-video-Hall'; opts = getdata(dataformat); time_getdata = toc; fprintf('%f seconds to get data ! \n', time_getdata); %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%% Call ALM to solve the problem tic; out_ALM = ALM_SADAL_smoothed(opts.D,opts); time_ALM = toc; fprintf('*******************************************************************\n'); %%%%%%%%% print stats %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% fprintf('*******************************************************************\n'); fprintf('ALM : iter: %d, StopCrit: %3.2e, time: %f\n', ... out_ALM.iter, out_ALM.StopCrit, time_ALM); % plot the images D = opts.D; X = out_ALM.X; Y = out_ALM.Y; imn1 = 144; imn2 = 176; ind1 = 30; ind2 = 96; ind3 = 159; subplot(3,3,1); imshow(reshape(D(:,ind1),imn1,imn2),[]); axis off; title('Video') subplot(3,3,2); imshow(reshape(X(:,ind1),imn1,imn2),[]); axis off; title('Background') subplot(3,3,3); imshow((reshape(Y(:,ind1),imn1,imn2)),[]); axis off; title('Foreground') subplot(3,3,4); imshow(reshape(D(:,ind2),imn1,imn2),[]); axis off; subplot(3,3,5); imshow(reshape(X(:,ind2),imn1,imn2),[]); axis off; subplot(3,3,6); imshow((reshape(Y(:,ind2),imn1,imn2)),[]); axis off; subplot(3,3,7); imshow(reshape(D(:,ind3),imn1,imn2),[]); axis off; subplot(3,3,8); imshow(reshape(X(:,ind3),imn1,imn2),[]); axis off; subplot(3,3,9); imshow((reshape(Y(:,ind3),imn1,imn2)),[]); axis off;