Given an array of co-ordinates, for each value of x, how do I extract the smallest value of y?

Context:

I have the pixel co-ordinates of the white area of a binary image.

For each x co-ordinate (1:1024) I want to find the smallest value of y. I want to do this because I want to find the 1D upper line boundary co-ordinates of a white splodge that bisects the image (see attached).

e.g for

x y

1 1

1 2

1 3

2 3

2 2

3 1

3 2

3 3

I want to be able to extract (1,1), (2,2), and (3,1) automatically, into a new array.

Thanks!

Answer 1

[Ameer Hamza]

Try something like this

M = [
    1    1
    1    2
    1    3
    2    3
    2    2
    3    1
    3    2
    3    3];

x = M(:,1);
y = M(:,2);

out = [(1:max(x)).' splitapply(@min, y, x)]

Result

>> out
out =
     1     1
     2     2
     3     1

An alternative solution using accumarray

out = [(1:max(x)).' accumarray(x, y, [], @min)]

[Lydia Bradley]

Thanks a lot, elegant solution!

Answer 2

[KSSV]

LEt (x,y) be your points.

[c,ia,ib] = unique(x) ; 
n = length(c) ; 
iwant = zeros(n,2) ; 
for i = 1:n
    iwant(i,1) = c(i) ; 
    iwant(i,2) = min(y(ib==i)) ; 
end
iwant

Answer 3

[madhan ravi]

x = [1; 1; 1; 2; 2; 3; 3; 3];
y = [1; 2; 3; 3; 2; 1; 2; 3];
T = array2table([x, y], 'VariableNames', {'x', 'y'});
T1 = groupsummary(T, 'x', 'min')

T1 = 3x3 table
    x    GroupCount    min_y
    _    __________    _____

    1        3           1  
    2        2           2  
    3        3           1  

Wanted = T1{:,{'x', 'min_y'}}

Wanted = 3×2

1 1 2 2 3 1

●