find the closest point and taking the difference of y coordinates

let assume I have A matrix in which there are m points and B matrix in which n points are there

I have a point lets say m1(x,y) in A matrix ,I want to find its closest/correspondent point in B matrix and after that want to differentiate the y coordinates of m1 point and its closest point

then taking m2(x,y) of A matrix finding it closest point in B matrix and diffrentiate and so on for each point in A matrix i want closest point in B matrix and diffrentiate their repsective y coordinates

I have tries some function/code like

1)"k = dsearchn(X,XI)"

2)

A = rand(1,2);
B = rand(10,2);
%compute Euclidean distances:
distances = sqrt(sum(bsxfun(@minus, B, A).^2,2));
%find the smallest distance and use that as an index into B:
closest = B(find(distances==min(distances)),:);

but didnt get the desired result

I am using matlab2014a

[Adam]

[~,idx] = min( distances );
B(idx,:);

should give you what you want, more simply, though it ought to just be a simpler version of what you have there. You didn't say what is wrong with it though other than 'didn't get the desired result'. What did you get?

Answer 1

[Adam Danz]

I recommen using pdist2() to calculate a matrix of distances between all points in A and all points in B. See comments within the block of code below.

% produces fake data
A = randi(100,3,2);
B = randi(100,5,2);

distance  = pdist2(A,B); 
% distance(i,j) is the distance between A(i,:) and B(j,:)

To find the nearest point in B to A(m,:),

% for m = 2, the row in B closest to A(m,:) is "minIdx"
m = 2; 
[~, minIdx] = min(distance(m,:)); 

I didn't follow the second part of your quesiton but here's how to pull those two coordinates from A and B.

m2 = A(m,:)
n2 = B(minIdx,:)