how to find k-th nearest neighbor of a point

i have many rectangles in the image represented by its centriod. How can i calculate the 10th nearest neighbor rectangles (centroids) for each ?

Answer 1

[Image Analyst]

Here's a way to do it if you don't have the Stats toolbox.

% Create sample data.
x = rand(1,100);
y = rand(1,100);

% Now we can start.
% Find number of centroid points.
numPoints = length(x);
% Make an array to keep track of the index of the 10th closest
% and that 10th closest point's index in the array.
tenthClosest = zeros(numPoints, 2);
for k = 1 : numPoints
  % Compute the distances of kth point to every other point (including itself).
  distances = sqrt((x(k)-x).^2 + (y(k) - y).^2);
  % Sort them so we can get the 10th distance at index 11
  % since there will be one point at 0 which is the distance of the point to itself which we don't care about.
  [sortedDistances, sortOrder] = sort(distances, 'Ascend');
  tenthClosest(k, 1) = sortedDistances(11);
  tenthClosest(k, 2) = sortOrder(11);
end
% Print to command window
tenthClosest

[MA-Winlab]

@Image Analyst

Would you please see this

Answer 2

[KSSV]

doc knnsearch