Find square cell centres in a noisy checker board/chess board image using pattern detection

Dear all,

I have a checker board/chess board pattern detection problem. I am interested in knowing the centres of the squares in my image frame.

I have tried the following approach:

clc;
close all;
clear;
% Load attached image (Extract the tiff file from the zip file :-) )
I = imread('23136695_Tilt2.tiff');
% Increase contrast of the image
I = imadjust(I);
% Crop the right half of the image for convenience and avoiding the dark 
% blob on the left half of the image
I = I(:, 695:end);

% The code below is taken from Matlab help sections on detecting edges and 
% examples provided by matlab users

BW = edge(I, 'Canny');
figure;
imshowpair(I,BW,'montage')

% Calculate the hough transform of the edge detected binarised image above.

figure;
[H,T,R] = hough(BW);
imshow(H,[],'XData',T,'YData',R,...
            'InitialMagnification','fit');
xlabel('\theta'), ylabel('\rho');
axis on, axis normal, hold on;

% Plot the hough transform peaks by thresholding

P  = houghpeaks(H,40,'threshold',ceil(0.3*max(H(:))));
x = T(P(:,2)); y = R(P(:,1));
plot(x,y,'s','color','white');

% Attempt to plot the lines (vertical and horizontal) running between the 
% square cells such that cell centres can be extracted.

lines = houghlines(BW,T,R,P,'FillGap',5,'MinLength',7);
figure, imshow(I), hold on
max_len = 0;
for k = 1:length(lines)
   xy = [lines(k).point1; lines(k).point2];
   plot(xy(:,1),xy(:,2),'LineWidth',2,'Color','green');

   % Plot beginnings and ends of lines
   plot(xy(1,1),xy(1,2),'x','LineWidth',2,'Color','yellow');
   plot(xy(2,1),xy(2,2),'x','LineWidth',2,'Color','red');

   % Determine the endpoints of the longest line segment
   len = norm(lines(k).point1 - lines(k).point2);
   if ( len > max_len)
      max_len = len;
      xy_long = xy;
   end
end

As you can see, this is not the desired result. I would need the complete grid detected. Is there a simple, computationally non-intensive (as a last resort, it is still OK to be laborious task as cell centre detection will only be done once) way out to solve this.

Answer 1

[Image Analyst]

What I'd do is to get the sum or mean of the gray level along each direction and look for peaks.

verticalProfile = mean(grayImage, 2);
figure
subplot(2, 1, 1);
plot(verticalProfile, 'b-', 'LineWidth', 2);
grid on;
horizontalProfile = mean(grayImage, 1);
subplot(2, 1, 2);
plot(horizontalProfile, 'b-', 'LineWidth', 2);
grid on;

Then call findpeaks(). Give that a try and report back here to me.