search

qianngli / ERCSR

Exploring the Relationship between 2D/3D Convolution for Hyperspectral Image Super-Resolution
20 stars 2 forks source link

Pivia Center preprocessing #1

Open chengyingsong opened 2 years ago

chengyingsong commented 2 years ago

It seems Pivia Center dataset preprocessing program is not exists. Can you provide it ?

qianngli commented 2 years ago

Eason @.***> 于2022年4月14日周四 11:50写道：

It seems Pivia Center dataset preprocessing program is not exists. Can you provide it ?

— Reply to this email directly, view it on GitHub https://github.com/qianngli/ERCSR/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANJ3LI7QIY2DG5POLZMTAFTVE6IXHANCNFSM5TMQXKNA . You are receiving this because you are subscribed to this thread.Message ID: @.***>

clc clear close all

%% define hyperparameters Band = 102; patchSize = 32; randomNumber = 108; upscale_factor = 8; data_type = 'PaviaCentre'; P = 0; global count count = 0;

imagePatch = patchSize*upscale_factor; scales = [1.0, 0.75];

%% bulid upscale folder savePath=['E:\PaviaCentre\8\']; if ~exist(savePath, 'dir') mkdir(savePath) end

%% srPath = 'E:/数据集/HyperSR/PaviaCentre/'; fileFolder=fullfile(srPath); dirOutput=dir(fullfile(fileFolder,'*.mat')); fileNames={dirOutput.name}'; length(fileNames)

for i = 1:length(fileNames) name = char(fileNames(i)); disp(['----:',data_type,'----upscale_factor:',num2str(upscale_factor),'----deal with:',num2str(i),'----name:',name]); data_path = [srPath, '/', name]; load(data_path) %% normalization radHeight = size(pavia, 1); radWidth = size(pavia, 2); t = pavia / (1.0*max(max(max(pavia)))); t = t(1:876,:,:);

clear pavia;

%% for j = 1:randomNumber for sc = 1:length(scales) newt = imresize(t, scales(sc));

    x_random = randperm(size(newt,1) - imagePatch, randomNumber);
    y_random = randperm(size(newt,2) - imagePatch, randomNumber);
    hrImage = newt(x_random(j):x_random(j)+imagePatch-1, y_random(j):y_random(j)+imagePatch-1, :);

    label = hrImage;   
    lr = data_augment(label, upscale_factor, savePath);
    sizeLR = size(lr);
    band_mean(j*sc,:) = mean(reshape(lr,[sizeLR(1), sizeLR(2)*sizeLR(3)]), 2);  

     if (rand(1)> P)
         label = imrotate(hrImage,180);  
         lr = data_augment(label, upscale_factor, savePath);
     end
    if (rand(1)> P)
        label = imrotate(hrImage,90);  
        lr = data_augment(label, upscale_factor, savePath);
    end
      if (rand(1)> P)
         label = imrotate(hrImage,270); 
         lr = data_augment(label, upscale_factor, savePath);
      end
    if (rand(1)> P)
        label = flipdim(hrImage,1); 
        lr = data_augment(label, upscale_factor, savePath);
    end
    if (rand(1)> P)
        label = flipdim(hrImage,2);  
         lr = data_augment(label, upscale_factor, savePath);
    end  

    clear x_random;
    clear y_random;
    clear newt;
end

end clear t; end

clc clear close all

upscale = 4; dataset = 'PaviaCentre';

%% bulid upscale folder % savePath=['\10.168.63.225\Home Share\liqiang\hyperSR\test\',dataset,'\',num2str(upscale)]; savePath=['E:/',dataset,'/',num2str(upscale),'/']; if ~exist(savePath, 'dir') mkdir(savePath) end

%% srPath = 'E:/数据集/HyperSR/PaviaCentre/'; fileFolder=fullfile(srPath); dirOutput=dir(fullfile(fileFolder,'*.mat')); fileNames={dirOutput.name}'; length(fileNames) aver_PSNR = 0; aver_SAM = 0; aver_SSIM = 0; count = 0;

for i = 1:length(fileNames) name = char(fileNames(i)); disp(['----:',dataset,'----upscale_factor:',num2str(upscale),'----deal with:',num2str(i),'----name:',name]); data_path = [srPath, '/', name]; load(data_path)

%% normalization radHeight = size(pavia, 1); radWidth = size(pavia, 2); t = pavia / (1.0*max(max(max(pavia))));

one = t(878:end,1:357,:); two = t(878:end,358:end,:);

HR = modcrop(one, upscale); LR = imresize(HR,1/upscale,'bicubic'); %LR
SR = imresize(LR,upscale,'bicubic'); save([savePath,'/1.mat'], 'HR', 'LR') n_PSNR = PSNR(SR, HR); n_SAM = SAM(SR, HR); n_SSIM = SSIM(SR, HR); aver_PSNR = aver_PSNR + n_PSNR; aver_SAM = aver_SAM + n_SAM; aver_SSIM = aver_SSIM + n_SSIM;
disp(['upscale_factor:',num2str(upscale),' ',num2str(1),' PSNR:',num2str(n_PSNR),' SAM:',num2str(n_SAM),... ' SSIM:',num2str(n_SSIM),' Name:',name]);

HR = modcrop(two, upscale); LR = imresize(HR,1/upscale,'bicubic'); %LR
SR = imresize(LR,upscale,'bicubic'); save([savePath,'/2.mat'], 'HR', 'LR') n_PSNR = PSNR(SR, HR); n_SAM = SAM(SR, HR); n_SSIM = SSIM(SR, HR); aver_PSNR = aver_PSNR + n_PSNR; aver_SAM = aver_SAM + n_SAM; aver_SSIM = aver_SSIM + n_SSIM;
disp(['upscale_factor:',num2str(upscale),' ',num2str(2),' PSNR:',num2str(n_PSNR),' SAM:',num2str(n_SAM),... ' SSIM:',num2str(n_SSIM),' Name:',name]);

end disp(['-----averPSNR:',num2str(aver_PSNR/2),'------averSAM:',num2str(aver_SAM/2),... '-----SSIM:',num2str(aver_SSIM/2)]);

function sam2 = SAM(imagery1,imagery2)

tmp = (sum(imagery1.imagery2, 3) + eps) ... ./ (sqrt(sum(imagery1.^2, 3)) + eps) ./ (sqrt(sum(imagery2.^2, 3)) + eps); sam = mean2(real(acos(tmp))); % this ignores the pixels that have zero norm (all values zero - no color) ind = ~isnan(sam); sam2 = mean(sam(ind))180/pi; end

function total_ssim = SSIM(SR, HR, K, window, L) total_ssim = 0; for i = 1 : size(SR, 3) img1 = SR(:,:,i); img2 = HR(:,:,i);
%======================================================================== %SSIM Index, Version 1.0 %Copyright(c) 2003 Zhou Wang %All Rights Reserved. % %The author is with Howard Hughes Medical Institute, and Laboratory %for Computational Vision at Center for Neural Science and Courant %Institute of Mathematical Sciences, New York University. % %---------------------------------------------------------------------- %Permission to use, copy, or modify this software and its documentation %for educational and research purposes only and without fee is hereby %granted, provided that this copyright notice and the original authors' %names appear on all copies and supporting documentation. This program %shall not be used, rewritten, or adapted as the basis of a commercial %software or hardware product without first obtaining permission of the %authors. The authors make no representations about the suitability of %this software for any purpose. It is provided "as is" without express %or implied warranty. %---------------------------------------------------------------------- % %This is an implementation of the algorithm for calculating the %Structural SIMilarity (SSIM) index between two images. Please refer %to the following paper: % %Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, "Image %quality assessment: From error measurement to structural similarity" %IEEE Transactios on Image Processing, vol. 13, no. 1, Jan. 2004. % %Kindly report any suggestions or corrections to @.*** % %---------------------------------------------------------------------- % %Input : (1) img1: the first image being compared % (2) img2: the second image being compared % (3) K: constants in the SSIM index formula (see the above % reference). defualt value: K = [0.01 0.03] % (4) window: local window for statistics (see the above % reference). default widnow is Gaussian given by % window = fspecial('gaussian', 11, 1.5); % (5) L: dynamic range of the images. default: L = 255 % %Output: (1) mssim: the mean SSIM index value between 2 images. % If one of the images being compared is regarded as % perfect quality, then mssim can be considered as the % quality measure of the other image. % If img1 = img2, then mssim = 1. % (2) ssim_map: the SSIM index map of the test image. The map % has a smaller size than the input images. The actual size: % size(img1) - size(window) + 1. % %Default Usage: % Given 2 test images img1 and img2, whose dynamic range is 0-255 % % [mssim ssim_map] = ssim_index(img1, img2); % %Advanced Usage: % User defined parameters. For example % % K = [0.05 0.05]; % window = ones(8); % L = 100; % [mssim ssim_map] = ssim_index(img1, img2, K, window, L); % %See the results: % % mssim %Gives the mssim value % imshow(max(0, ssim_map).^4) %Shows the SSIM index map % %======================================================================== if (nargin < 2 || nargin > 5) ssim_index = -Inf; ssim_map = -Inf; return; end

if (size(img1) ~= size(img2)) ssim_index = -Inf; ssim_map = -Inf; return; end

[M N] = size(img1);

if (nargin == 2) if ((M < 11) || (N < 11)) ssim_index = -Inf; ssim_map = -Inf; return end window = fspecial('gaussian', 11, 1.5); % K(1) = 0.01; % default settings K(2) = 0.03; % L = 1; % end

if (nargin == 3) if ((M < 11) || (N < 11)) ssim_index = -Inf; ssim_map = -Inf; return end window = fspecial('gaussian', 11, 1.5); L = 255; if (length(K) == 2) if (K(1) < 0 || K(2) < 0) ssim_index = -Inf; ssim_map = -Inf; return; end else ssim_index = -Inf; ssim_map = -Inf; return; end end

if (nargin == 4) [H W] = size(window); if ((H*W) < 4 || (H > M) || (W > N)) ssim_index = -Inf; ssim_map = -Inf; return end L = 255; if (length(K) == 2) if (K(1) < 0 || K(2) < 0) ssim_index = -Inf; ssim_map = -Inf; return; end else ssim_index = -Inf; ssim_map = -Inf; return; end end

if (nargin == 5) [H W] = size(window); if ((H*W) < 4 || (H > M) || (W > N)) ssim_index = -Inf; ssim_map = -Inf; return end if (length(K) == 2) if (K(1) < 0 || K(2) < 0) ssim_index = -Inf; ssim_map = -Inf; return; end else ssim_index = -Inf; ssim_map = -Inf; return; end end

C1 = (K(1)L)^2; C2 = (K(2)L)^2; window = window/sum(sum(window)); img1 = double(img1); img2 = double(img2);

mu1 = filter2(window, img1, 'valid'); mu2 = filter2(window, img2, 'valid'); mu1_sq = mu1.mu1; mu2_sq = mu2.mu2; mu1_mu2 = mu1.mu2; sigma1_sq = filter2(window, img1.img1, 'valid') - mu1_sq; sigma2_sq = filter2(window, img2.img2, 'valid') - mu2_sq; sigma12 = filter2(window, img1.img2, 'valid') - mu1_mu2;

if (C1 > 0 & C2 > 0) ssim_map = ((2mu1_mu2 + C1).(2sigma12 + C2))./((mu1_sq + mu2_sq + C1).(sigma1_sq + sigma2_sq + C2)); else numerator1 = 2mu1_mu2 + C1; numerator2 = 2sigma12 + C2; denominator1 = mu1_sq + mu2_sq + C1; denominator2 = sigma1_sq + sigma2_sq + C2; ssim_map = ones(size(mu1)); index = (denominator1.denominator2 > 0); ssim_map(index) = (numerator1(index).numerator2(index))./(denominator1(index).*denominator2(index)); index = (denominator1 ~= 0) & (denominator2 == 0); ssim_map(index) = numerator1(index)./denominator1(index); end

mssim = mean2(ssim_map); total_ssim = total_ssim + mssim; end total_ssim = total_ssim/size(SR, 3); end function single_psnr = PSNR(HR_image, SR_image) HR_image = reshape(HR_image, size(HR_image,1)size(HR_image, 2), size(HR_image,3));
SR_image = reshape(SR_image, size(SR_image,1)size(SR_image, 2), size(SR_image,3));
L = 1; width = size(HR_image,2); height = size(HR_image,1); if( width ~= size(SR_image,2) || height ~= size(SR_image,1) ) disp('Please check the input image have the same size'); return end

HR_image = double(HR_image);
SR_image = double(SR_image);
MES = mean((HR_image(:) - SR_image(:)).^2);
single_psnr=10*log10(L^2/MES);

end

jerryniu0624 commented 2 months ago

which is the script for training and testing? I assume line 1-88 is for training. Here is an error: image

jerryniu0624 commented 2 months ago

why do this script have the metrics computation? this script is for input data generation, right?