How do I use the find function to track every last point that is black, so as to create a matrix of that border between white a black image. See Below

I used the find function to find a single point, the row at which the image is black, in the first column. But I want to do this for every column, so as to create a defined line of every row and and column, instead of a single point

clear;clc;

img1 = imread('s1.png');

img2 = imread('s2.png');

img1BW = rgb2gray(img1);

img2BW = rgb2gray(img2);

counter = 0;

for k = 1:length(img1BW);

counter = counter+1;

end

for k = 1:length(img2BW);

[r2,c] = find(img2BW(:,1:end),1,'last');

counter = counter+1;

end

Border1 = [r1,c]

Border2 = [r2,c]

final = abs(r2-r1);

fprintf('The displacement of the black border from image 1 and image 2 is : %1.000f pixels \n',final);

[yanqi liu]

yes，sir，is it to find the white rectangle location，then can use the top、bottom、left、right to segment it

[alex contreras]

I want to create an array/matrix of every point that is on that line that dinstinguishes black from white.

[alex contreras]

from this I can subtract every one those points from either image to find the difference.

Answer 1

[Image Analyst]

Try this. It assumes the top part is all white, and the only black is a big region along the bottom, which doesn't have to be perfectly rectangular.

[rows, columns] = size(img1BW);
topEdgeOfBlack = zeros(1, columns);
% Scan column-by-column finding out where (what row) the first black pixel 
% in each column appears.
for col = 1 : columns
    topEdgeOfBlack(col) = find(img1BW(:, col) == 0, 1, 'first');
end
hold on;
x = 1 : columns;
plot(x, topEdgeOfBlack, 'r-', 'LineWidth', 3);
hold off;

Alternatively you can find the last white pixel in each column like this:

[rows, columns] = size(img1BW);
bottomEdgeOfWhite = zeros(1, columns);
% Scan column-by-column finding out where (what row) the last white pixel 
% in each column appears.
for col = 1 : columns
    bottomEdgeOfWhite(col) = find(img1BW(:, col) > 0, 1, 'last');
end
hold on;
x = 1 : columns;
plot(x, bottomEdgeOfWhite, 'r-', 'LineWidth', 3);
hold off;

[alex contreras]

wont the second option result in a single row?, I want it to find every row and column that has the last white pixel. In future projects, the images I will be using won't be flat distinct lines.

[Image Analyst]

@alex contreras, no. Why would all elements of bottomEdgeOfWhite be the same row? They won't, unless you have a perfectly flat bottom to the white region. If the bottom edge varies/waivers/wanders, then bottomEdgeOfWhite will have a variety of rows that follow the bottom.

[alex contreras]

Sorry, but what am I doing wrong? I replaced the sample image with a new one where the edge varies(Bottom left) and bottomEdgeOfWhite returns the same repeated element.

[Image Analyst]

You forgot to attach s3.png and your code. And don't make me type in those 15 lines of code (because I won't). Make it easy for me to help you and I will.

[alex contreras]

clear;clc;
img1 = imread('s3.png');
img1BW = rgb2gray(img1);

[rows, columns] = size(img1BW);
bottomEdgeOfWhite = zeros(1, columns);
% Scan column-by-column finding out where (what row) the last white pixel 
% in each column appears.
for col = 1 : columns
    bottomEdgeOfWhite = find(img1BW > 0, 1, 'last')
end
hold on;
x = 1 : columns;
plot(x, bottomEdgeOfWhite, 'r-', 'LineWidth', 3);
hold off;

[alex contreras]

this is my first time posting for help... sorry for not uploading it correctly.

[alex contreras]

clear;clc;
img1 = imread('s3.png');
img1BW = rgb2gray(img1);

[rows, columns] = size(img1BW);
bottomEdgeOfWhite = zeros(1, columns);
% Scan column-by-column finding out where (what row) the last white pixel 
% in each column appears.
for col = 1 : columns
    bottomEdgeOfWhite = find(img1BW > 0, 1, 'last')
end
hold on;
x = 1 : columns;
plot(x, bottomEdgeOfWhite, 'r-', 'LineWidth', 3);
hold off;

[Image Analyst]

Try this:

img1 = imread('s3.png');
img1BW = rgb2gray(img1);
imshow(img1BW);
axis('on', 'image');
impixelinfo;
[rows, columns] = size(img1BW);
bottomEdgeOfWhite = zeros(1, columns);
% Scan column-by-column finding out where (what row) the last white pixel 
% in each column appears.
for col = 1 : columns
    bottomEdgeOfWhite(col) = find(img1BW(:, col) > 0, 1, 'last');
end
hold on;
x = 1 : columns;
plot(x, bottomEdgeOfWhite, 'r-', 'LineWidth', 3);
hold off;

[alex contreras]

Thank you very much!! I really appreciate it.

Answer 2

[KSSV]

I = imread(myimage) ; 
I1 = imbinarize(rgb2gray(I)) ;
[y,x] = find(I1) ;
idx = boundary(I1) ;
plot(x(idx),y(idx))

You can also use regionprops. Have a look on the function.

[alex contreras]

I am getting this error:

Error using boundary>sanityCheckPoints (line 172)

The input points must be 2D or 3D coordinates in numpoints-by-ndim format.

Error in boundary>sanityCheckInput (line 117)

sanityCheckPoints(P);

Error in boundary (line 54)

[P, S] = sanityCheckInput(varargin{:});