how to execute for loop differently?...

In my code, I'm extracting and comparing images from the folder and the user is cropped at the figure. whether user cropped images and folder existing image features are same or not . when the user cropped and the existing folder images comparison is fine for the first time and when compared for the 2nd user cropped to already existing image folder is now comparing the first folder image comparison again. I don't need that. when comparison comes, user cropped first image and existing folder first image compare and again compare with the user cropped 2nd image and existing folder 2nd comparison I need. In my code, user cropped first image comparison with already existing folder 1st image and comparing the user cropped 2nd image and comparing existing folder 1st and 2nd image so on... Please help me for my project.

    function percent_match(uname1,imgs)
  disp(uname1);
  myDir = 'user_images/';
  myDir=strcat(myDir,uname1)
  myDir=strcat(myDir,'/');
  jpgFiles = dir([myDir '*.jpg']);
  numFiles = length(jpgFiles);
  mydata = cell(1,numFiles);

% mydata = zeros(numFiles);
for k = 1:numFiles
    mydata{k} = imread([myDir jpgFiles(k).name]);
    mydata{k}=rgb2gray(mydata{k});
    imgs=rgb2gray(imgs);
   mydataPoints=detectSURFFeatures(mydata{k});
   strong1 = mydataPoints.selectStrongest(50);
   imgsPoints=detectSURFFeatures(imgs);
   strong2 = imgsPoints.selectStrongest(50);
   [mydataFeatures,mydataPoints1]=extractFeatures(mydata{k},strong1);
[imgsFeatures,imgsPoints]=extractFeatures(imgs,strong2);
mydataPairs=matchFeatures(mydataFeatures,imgsFeatures);
matchedmydataPoints = mydataPoints(mydataPairs(:, 1), :);
matchedimgsPoints = imgsPoints(mydataPairs(:, 2), :);
[tform, inlierBoxPoints, inlierScenePoints] = ...
    estimateGeometricTransform(matchedmydataPoints, matchedimgsPoints, 'affine');
    boxPolygon = [1, 1;...                           % top-left
        size(mydata{k}, 2), 1;...                 % top-right
        size(mydata{k}, 2), size(mydata{k}, 1);... % bottom-right
        1, size(mydata{k}, 1);...                 % bottom-left
        1, 1];     
        newBoxPolygon = transformPointsForward(tform, boxPolygon);
        %figure;
%imshow(imgs);
%hold on;
%line(newBoxPolygon(:, 1), newBoxPolygon(:, 2), 'Color', 'y');
%title('Detected Box');
 numPairs = length(mydataPairs); %the number of pairs
     percentage  = (numPairs*2);
 if percentage >75
     disp('we have this');
     disp(percentage);
 else
     disp('we dont have this')
 end
   break;
  end
 end

[KSSV]

Jesus...1st image, 2nd image, cropping user.....question is very confusing..needs lots of focus to understand. I quit.