How to handle 100 binary images with a single matrix?

My folder consists of 100 binary images. I want to divide each of these images into some parts and want to calculate area of each block and want to store it in a matrix. Later that matrix I want to map to excel sheet I mean to say that I want to write these values to excel sheet. How to implement this repeated process so that 100 images each blocks area value will be stored in a matrix.

[Guillaume]

Your question is really lacking in details making it hard to answer. What bit of your process is giving you trouble?

[Zara Khan]

I am not able to store each blocks areas to a matrix then it in excel sheet. As I have 100 images. And I am diving each of theSe images into 4 or 8 parts. So there will be either 400 or 800 blocks.I want to store these areas in a matrix then it in a excel file.

Answer 1

[Image Analyst]

Since your blocks are just simply quadrants, as I can see from your code attempt, do it like this:

folder = 'D:\data_set1';
% folder = pwd;
filepattern = fullfile(folder, '*.png');
srcFiles = dir(filepattern);
numImages = length(srcFiles);
allAreas = zeros(numImages, 4);
areaFractions = zeros(numImages, 4);
for k = 1 : numImages
  fullFileName = fullfile(folder, srcFiles(k).name);
  grayImage = imread(fullFileName);
  % Get the dimensions of the image.
  % numberOfColorChannels should be = 1 for a gray scale image, and 3 for an RGB color image.
  [rows, columns, numberOfColorChannels] = size(grayImage);
  if numberOfColorChannels > 1
    % It's not really gray scale like we expected - it's color.
    % Use weighted sum of ALL channels to create a gray scale image.
    grayImage = rgb2gray(grayImage);
  end
  % Display original image.
  subplot(1, 2, 1);
  imshow(grayImage);
  title(srcFiles(k).name, 'FontSize', 15, 'Interpreter', 'none');
  
  % Make sure the image is uint8 or uint16
  if ~isinteger(grayImage);
    continue;
  end
  
  % Threshold the image.
  binaryImage = imbinarize(grayImage);
  % Make sure there is only one blob in the image.
  binaryImage = bwareafilt(binaryImage, 1);
  % Display binary image.
  subplot(1, 2, 2);
  imshow(binaryImage);
  title('Binary Image', 'FontSize', 15, 'Interpreter', 'none');
  drawnow;
  % Get the centroid and the bonding box.
  % Not sure why since we don't use them so I'll comment it out.
%   props = regionprops(binaryImage, 'Centroid', 'BoundingBox');

  % Determine half way point.  Need to round in case image has odd number of rows or columns.
  rowMid = round(rows / 2);
  colMid = round(columns / 2);
  
  % Divide the binary image up into quadrants.
  q1 = binaryImage(1:rowMid, 1:colMid);
  q2 = binaryImage(1:rowMid, (colMid+1):end);
  q3 = binaryImage((rowMid+1):end, colMid);
  q4 = binaryImage((rowMid+1):end, (colMid+1):end);
  % Get the foreground area in each quadrant.
  area1 = sum(q1(:));
  area2 = sum(q2(:));
  area3 = sum(q3(:));
  area4 = sum(q4(:));
  allAreas(k, :) = [area1, area2, area3, area4];
  areaFractions(k, :) = [area1, area2, area3, area4] ./ [numel(q1), numel(q2), numel(q3), numel(q4)];
end
% Report to the command window.
allAreas
areaFractions

[Zara Khan]

Image Analyst: Thank you so much. It has helped me a lot. This is the exact solution of my problem.

Answer 2

[Image Analyst]

Use blockproc() in a loop over all images.

[Zara Khan]

But not able to store all blocks areas in a single matrix

[Image Analyst]

For 4 areas, the area of each quadrant is numel(binaryImage)/4.

If you divide it into 8, the area is numel(binaryImage)/8 pixels.

For one image, you could of course store all the quadrant areas into one matrix.

areas4 = numel(binaryImage)/4 * ones(1, 4);
areas8 = numel(binaryImage)/8 * ones(1, 8);

If you want the foreground area of the binary image rather than the area of the image, you'll have to sum each quadrant

q1 = binaryImage(1:rows/2, 1:columns/2);
q2 = binaryImage(1:rows/2, columns/2:end);
q3 = binaryImage(rows/2:end, 1:columns/2);
q4 = binaryImage(rows/2:end, columns/2:end);
area1 = sum(q1(:));
area2 = sum(q2(:));
area3 = sum(q3(:));
area4 = sum(q4(:));

[Zara Khan]

i
srcFiles = dir('D:\data_set1\*.png'); 
for i = 1 : length(srcFiles)
  filename = strcat('D:\data_set1\',srcFiles(i).name);
  a = imread(filename);
  bwimg =bwareafilt(~a, 1); 
  props = regionprops(bwimg,'Centroid','BoundingBox');
assert(numel(props) == 1, 'more than one object found in image '); 
[rows columns]=size(bwimg);
areas4 = numel(bwimg)/4 * ones(1, 4);
q1 = bwimg(1:rows/2, 1:columns/2);
q2 = bwimg(1:rows/2, columns/2:end);
q3 = bwimg(rows/2:end, 1:columns/2);
q4 = bwimg(rows/2:end, columns/2:end);
area1 = sum(q1(:));
area2 = sum(q2(:));
area3 = sum(q3(:));
area4 = sum(q4(:));
end

Image Analyst: When I am doing this for a set of 100 images from a folder then all these areas are not storing in a single matrix. How to do that ? As I am new many things are not still clear to me. After storing these values in a single matrix I want to write them in a excel file.

[Zara Khan]

I have written my code here. Please anyone help me in doing the part