1.仿真預覽

2.部分核心代碼

% % Geometric Blind Deconvolution script % % Copyright 2006 Paolo Favaro (p.favaro@hw.ac.uk) % % School of Engineering and Physical Sciences % Heriot-Watt University, Edinburgh, UK % % Last revision: May 2006 % % This program can be used only for research purposes. % This program is distributed WITHOUT ANY WARRANTY; % without even the implied warranty of MERCHANTABILITY % or FITNESS FOR A PARTICULAR PURPOSE.clear all close all% set to 1 to see plots, 0 otherwise plotfigures = 1;Yes = 1; No = 0; fprintf('Shape from Defocus (Least Squares)\n'); notValid = Yes; while notValidmethod = input(['Which method do you want to test?'...'\nChoose generate, learn, overlap: '],'s'); notValid = ~strcmp(method,'generate')&...~strcmp(method,'learn')&...~strcmp(method,'overlap'); end %%%%%%%%%%%%%%%%%%%%% % compute operators % %%%%%%%%%%%%%%%%%%%%% % ni is the square patch size % the larger ni the slower the algorithm notValid = Yes; while notValidni = input(['Choose the patch size '...'\n3(fast&noisy),5,7,9(slow&smooth): '],'s');notValid = ~(ni=='3')&...~(ni=='5')&...~(ni=='7')&...~(ni=='9'); end ni = str2num(ni); % fix rank of operators ranks = ni*ni; % operator ranks if strcmp(method,'generate')% compute operators when PSF is knownHp = generate_operators(ni,ranks); elseif strcmp(method,'learn')% compute operators when PSF is unknown % learning approach; non overlapping patchesHp = learn_operators(ni,ranks); else% compute operators when PSF is unknown% learning approach; overlapping patchesHp = learn_operators_overlap(ni,ranks); end %%%%%%%%%%%%%%% % load images % %%%%%%%%%%%%%%% load DataSet%%%%%%%%%%%%%%%%%% % Estimate Depth % %%%%%%%%%%%%%%%%%% Depth = estimate_depth(I1,I2,Hp,plotfigures);% optional smoothing FilteredDepth = mediansmoothing(Depth,plotfigures);return

C5

總結

以上是生活随笔為你收集整理的通过MATLAB提取图像的深度信息的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。