The function returns N-D logical array (same size as the input array) with the n largest connected-components indices set.
Inputs:
-------
Xin - input N-D array. Required.
conn - connectivity definition, can be either a scalar or connectivity array, with same number of dimensions as the input array. If empty or omitted , 'conn' will be set to scalar 3^ndims(Xin)-1.
n - Number of returned largest connected-components. If omitted, 'n' will be set to 1. If 'n' is greater than N, the actual number of connected-components in the array, N will be used instead.
Outputs:
--------
Xout - Output N-D array (same size as Xin), with only the n largest connected-components indices set.
rp - vector of length n of structures containing information about the size of the connected-components and their indices, arranged in descending order of size.
Example (2D):
I = rand(100,100) > 0.75;
[J rp] = getLargestCc(I,[],3);
I is a random 2D "image". J is a copy of I with only the 3 largest connected-components set.
Despite the fact that the example is for a very simple 2D image, the code applies to N-D arrays as well.
Comments and questions to: ran.shadmi@gmail.com. |