Batch processing images for downsampling

Question

Cemre 2023-6-13

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1982349-batch-processing-images-for-downsampling

评论： Mathieu NOE 2023-6-14

Hi,

I would like to perform downsampling for like over 700 images. It is pretty time-consuming to do it one by one. How can I achieve it at one time?

Thanks a lot.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Mathieu NOE 2023-6-13

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1982349-batch-processing-images-for-downsampling#answer_1254859

在 MATLAB Online 中打开

hello

see example below - adapt to your own needs

%%  Initalize the data 
dataDir= fullfile(pwd); % select appropriate directory
exts = {'.jpg','.png','.tif'}; % choose valid file extensions like {'.jpg', '.png'}
 
resize_size = 100;      % pixels size for output img 
 
imds = imageDatastore(fullfile(dataDir),...
'IncludeSubfolders',true,'FileExtensions',exts,'LabelSource','foldernames');
countEachLabel(imds);
 
numImages = numel(imds.Files); 
for i = 1:numImages
    img = readimage(imds, i);
    [m,n,p] = size(img);
    % compute scale factor (same on both  dimensions)
    scale_factor = min(resize_size/m,resize_size/n);
    img3= imresize(img, scale_factor);
    figure(i),
    img4= imshow(img3, 'InitialMagnification', 800);
     drawnow;
    Train{i} = (img3);         %output image stored in cell
end

8 个评论
显示 6更早的评论隐藏 6更早的评论

Mathieu NOE 2023-6-13

在 MATLAB Online 中打开

another example

% select appropriate options below and adapt code to your specific needs ...
% files = dir('*.tif');      % select all tiff files in current directory
files = dir('handheld*.tif');      % select only tiff files with "handheld" in the filename
% resize options (sacle factor or pixels)
scale_factor = 0.5;
% select portion of the image
% 451*279*3. I will like to extract a portion with dimensions of 120*80*3
x_dim = 120;
y_dim = 80;
x_offset = 100; % please make sure the "corner" of the cropped image is correct
y_offset = 50;
% output directory 
out_dir = fullfile(pwd,'\out'); % from current directory to sub directory \out
% main loop
for ck = 1:length(files)
    Filename = files(ck).name;
    img = imread(Filename);
    
    % resize
    H{ck}= imresize(img, scale_factor);
    % select (crop)
    H{ck}= img(x_offset+(1:x_dim),y_offset+(1:y_dim),: );
    % plot (just to check)
    figure(ck), imagesc(H{ck});
    
    % insert your own code here (save for example)
    filename = ['out' num2str(ck) '.tiff'];
    imwrite(H{ck},fullfile(out_dir,filename));
    
end

Cemre 2023-6-14

在 MATLAB Online 中打开

% select appropriate options below and adapt code to your specific needs ...
files = dir('*.png');      % select all tiff files in current directory
%files = dir('handheld*.tif');      % select only tiff files with "handheld" in the filename
% resize options (sacle factor or pixels)
scale_factor = 0.2;
% select portion of the image
% 451*279*3. I will like to extract a portion with dimensions of 120*80*3
x_dim = 140;
y_dim = 140;
x_offset = 100; % please make sure the "corner" of the cropped image is correct
y_offset = 50;
% output directory 
out_dir = fullfile(pwd,'\out'); % from current directory to sub directory \out
% main loop
for ck = 1:length(files)
    Filename = files(ck).name;
    img = imread(Filename);
    
    % resize
    H{ck}= imresize(img, scale_factor);
    % select (crop)
    %H{ck}= img(x_offset+(1:x_dim),y_offset+(1:y_dim),: );
    % plot (just to check)
    figure(ck), imagesc(H{ck});
    
    % insert your own code here (save for example)
    filename = ['out' num2str(ck) '.tiff'];
    imwrite(H{ck},fullfile(out_dir,filename));
    
end

Also, size of the output image is 560x420. I would like it to be 140x140 and with bicubic interpolation as like in this example but it also gives me a different size.

% Load the original image
origImage = imread('Planet_20221104_1667.png');
% Set the desired scale factor to achieve 10m resolution
scaleFactor = 0.20;
% Resize the image using bicubic interpolation
downsampledImage = imresize(origImage, scaleFactor, 'bicubic');
% Save the downsampled image to a new file
imwrite(downsampledImage, 'Planet_20221104_0364_1.png');