getTrainingData.m

% only upright test images in directory
testFolder='pictures\'; %%%%% your test directory
files=dir(testFolder);

%get images
clear filenames
for i=3:size(files,1)
    filenames{i-2,1}=files(i).name;
end

clear trainingInput
clear trainingOutput
trainingInput = [];
trainingOutput = [];

%generate train data
for i=1:size(filenames,1)
    disp(i + "/" + size(filenames,1));    

    type=strsplit(filenames{i},'-');
    
    type=strsplit(type{2},'.');

    type=str2double(type{1});
    
    if(type == 1 | type == 2 )

        I=imread(strcat(testFolder,(filenames{i})));

        % Height = size(I, 1);
        % Width = size(I, 2);

        I = getTrainingImage(I);

        correctId=strsplit(filenames{i},'-');
        correctId=str2double(correctId{1});

        mkdir("trainingImages/"+correctId+"/")
        
        imwrite(I,"trainingImages/"+correctId+'/'+type+'.png')

        % trainingInput = cat(3, trainingInput, I);
        
        
        % trainingOutput=[trainingOutput, correctId];
        % size(trainingInput)
        % size(trainingOutput)

    end
end
% save('training','trainingInput','trainingOutput')

%This function is going to make the image lighter from the basic image
function [Image] = getTrainingImage(Image)
    % input : I (upright raw RGB image)
    % output : Image made for the training dataset
    %get

    % Image = getFaceCropped(Image);
    Image = imresize(Image, [224 224]);
    % Image = rgb2gray(Image);
    

end