How do I find the corner points of an mask

I have various mask and i want to find the exact four courner coordinates, but when some mask objects with an rotation and little odd shape comes i find it difficult to find the corner points.

I have attached the mask and pointed the points i want to find

Answer 1

[Mathieu NOE]

hello

let's try with detectHarrisFeatures : not really super efficient I admit , I will propose alternatives ...

im = imread('image.png');    % Load your binary mask image

% Convert the image to grayscale
gray_img = rgb2gray(im);

% Apply the Harris corner detector
corners = detectHarrisFeatures(gray_img,'MinQuality',0.5);

% Display the image with the detected corners
imshow(im); hold on;
plot(corners.selectStrongest(10));

●

[Mathieu NOE]

hello again

this is a (better) alternative based on boundary and its curvature computation

you may need to smooth the boundary curve , I used this for the job : smoothn - File Exchange - MATLAB Central

hope it helps

my result so far :

code :

A = imread('image.png'); 
% a bit of gym to convert to grayscale , that we will "binarize" with
% logical operation 
A = double(A);
% rgb2gray converts RGB values to grayscale values by forming a weighted sum of the R, G, and B components:
% 0.2989 * R + 0.5870 * G + 0.1140 * B 
A = 0.299 * A(:,:,1) + 0.587 * A(:,:,2) + 0.114 * A(:,:,3);
A = flipud(A); % to have image displayed with correct y direction
[y,x] = find(A>0.5*256); % find whte dots coordinates : threshold is set at 50% of 256
k = boundary(x,y,1); % find boundary
x_selec = x(k);
y_selec = y(k);
% smooth a bit the contour
z = smoothn({x_selec,y_selec},20);
x_selec = z{1};
y_selec = z{2};
xx_centroid = mean(x_selec);
yy_centroid = mean(y_selec);

% Plot curvature.
curvature = my_curvature(x_selec,y_selec);
[pks,locs] = findpeaks(curvature,'MinPeakHeight',max(curvature)/10);
xc = x_selec(locs);
yc = y_selec(locs);

% select the 4 points according to their x distance vs centroid
% dist = (xc-xx_centroid).^2+(yc-yy_centroid).^2;
dist = (xc-xx_centroid).^2;
[dist,ind] = sort(dist,'descend');
xc = xc(ind(1:4));
yc = yc(ind(1:4));

figure;
imagesc(A);
colormap('gray')
hold on
set(gca,'YDir','normal');
plot(x_selec,y_selec,'g');
plot(xc,yc,'dr');
grid on; axis equal
xlabel x
ylabel y
hold off

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function curvature = my_curvature(x,y)
%=====================================================
% Now run along the (x, y) soft star curve 
% and find the radius of curvature at each location.
numberOfPoints = length(x);
curvature = zeros(1, numberOfPoints);
    for t = 1 : numberOfPoints
      if t == 1
        index1 = numberOfPoints;
        index2 = t;
        index3 = t + 1;
      elseif t >= numberOfPoints
        index1 = t-1;
        index2 = t;
        index3 = 1;
      else
        index1 = t-1;
        index2 = t;
        index3 = t + 1;
      end
      % Get the 3 points.
      x1 = x(index1);
      y1 = y(index1);
      x2 = x(index2);
      y2 = y(index2);
      x3 = x(index3);
      y3 = y(index3);
      % Now call Roger's formula:
      % http://www.mathworks.com/matlabcentral/answers/57194#answer_69185
      curvature(t) = 2*((x2-x1).*(y3-y1)-(x3-x1).*(y2-y1)) ./ ...
      sqrt(((x2-x1).^2+(y2-y1).^2)*((x3-x1).^2+(y3-y1).^2)*((x3-x2).^2+(y3-y2).^2));
    end
end

[kunal]

Oh thankyou, I will try it and see if it works for my use cases.

Answer 2

[Matt J]

You could download pgonCorners,

https://www.mathworks.com/matlabcentral/fileexchange/74181-find-vertices-in-image-of-convex-polygon

BW = imbinarize( im2gray(imread('image.png')) );    

V=pgonCorners(BW,4);

imshow(BW,[]); hold on
plot(V(:,2),V(:,1),'r.',MarkerSize=30); hold off

●

[kunal]

Thankyou, I will try it

Answer 3

[Image Analyst]

Depending on the shapes you're dealing with, convhull may work for you.

Answer 4

[Image Analyst]

See my Answer in your duplicate question: https://www.mathworks.com/matlabcentral/answers/2177033-help-me-get-the-points-of-the-mask#answer_1564977