DatabaseLoad.m

%--------------------------------------------------------------------------
%   Title: Multi-Process Fusion
%   Author: Stephen Hausler
%   
%   Open Source Code, requires MATLAB with Neural Network Toolbox
%   Refer to LICENSES.txt for license to this source code and all 3rd party
%   licences.
%--------------------------------------------------------------------------

function [Template_array1,Template_array2,Template_array3,Template_array4,...
    totalImagesR,Template_count,Template_plot] = DatabaseLoad(varargin)

Template_count = 0;

if nargin == 15
    Video_option = varargin{1};
    Ref_folder = varargin{2};
    Ref_file_type = varargin{3};
    Imstart_R = varargin{4};
    Frame_skip = varargin{5};
    net = varargin{6};
    actLayer = varargin{7};
    SAD_resolution = varargin{8};
    SAD_patchSize = varargin{9};
    HOG_resolution = varargin{10};
    HOG_cellSize = varargin{11};
    CNN_resolution = varargin{12};
    Initial_crop = varargin{13};
    Normalise = varargin{14};
    finalImage_R = varargin{15};
else
    error('invalid number of inputs');
end

    Ref_file_type = strcat('*',Ref_file_type);
    fR = dir(fullfile(Ref_folder,Ref_file_type));

    Imcounter_R = Imstart_R;
    fR2 = struct2cell(fR);
    filesR = sort_nat(fR2(1,:));
    i = 1;
    
    while((Imcounter_R+1) <= finalImage_R)
        filenamesR{i} = filesR(Imcounter_R+1);
        Imcounter_R = Imcounter_R + Frame_skip;
        i=i+1;
    end
    
    totalImagesR = length(filenamesR);
    
    for i = 1:totalImagesR
        Im = imread(char(fullfile(fR(1).folder,filenamesR{i})));
        sz = size(Im);
        Im = Im((Initial_crop(1):(sz(1)-Initial_crop(2))),Initial_crop(3):(sz(2)-Initial_crop(4)),:);
        
        Im1 = imresize(Im,[CNN_resolution(2) CNN_resolution(1)],'lanczos3');  %for CNN
        Im2 = rgb2gray(Im);
        Im3 = imresize(Im2,[HOG_resolution(2) HOG_resolution(1)],'lanczos3'); %downsize for HOG
        Im4 = imresize(Im2,[SAD_resolution(2) SAD_resolution(1)],'lanczos3'); %downsize for SAD
        
        template1 = CNN_Create_Template(net,Im1,actLayer);      %CNN
        template2 = CNN_Create_Template_Dist(net,Im1,actLayer); %CNN-Dist
        template3 = extractHOGFeatures(Im3,'CellSize',[HOG_cellSize(1) HOG_cellSize(2)]); %HOG
        template4 = zeros(1,size(Im4,1)*size(Im4,2),'int8');    %SAD
        Im4P = patchNormalizeHMM(Im4,SAD_patchSize,0,0);
        template4(1,:) = Im4P(:);
        
        %Now store in template matrix:
        Template_count = Template_count + 1;
        Template_plot(Template_count,1) = i;
        Template_plot(Template_count,2) = Template_count;
        
        Template_array1(Template_count,:) = template1;
        Template_array2(1,:,Template_count) = template2(1,:);
        Template_array2(2,:,Template_count) = template2(2,:);  
        Template_array3(Template_count,:) = template3;
        Template_array4(Template_count,:) = template4;
    end
    if Normalise == 1
        fAv = mean(Template_array1,1);
        fSt = std(Template_array1,1);
        sz = size(Template_array1);

        %Now perform normalisation on every element in the template arrays.
        for j = 1:sz(2)
            if fSt(j) == 0 
                Template_array1(:,j) = 0;
            else
                Template_array1(:,j) = (Template_array1(:,j) - fAv(j))/fSt(j);
            end
        end   
    end
end