函数 

double  cv::kmeans ( InputArray data, int K, InputOutputArray bestLabels, TermCriteria criteria, int attempts, int flags, OutputArray centers= noArray ()) 
Finds centers of clusters and groups input samples around the clusters.
更多...


template<typename _Tp , class _EqPredicate >  
int  partition (const std::vector< _Tp > &_vec, std::vector< int > &labels, _EqPredicate predicate=_EqPredicate()) 
Splits an element set into equivalency classes.
更多...


double cv::kmeans  (  InputArray  data , 
int  K ,  
InputOutputArray  bestLabels ,  
TermCriteria  criteria ,  
int  attempts ,  
int  flags ,  
OutputArray 
centers
=
noArray
()


) 
Python:  

retval, bestLabels, centers  =  cv.kmeans(  data, K, bestLabels, criteria, attempts, flags[, centers]  ) 
#include <
opencv2/core.hpp
>
Finds centers of clusters and groups input samples around the clusters.
The function kmeans implements a kmeans algorithm that finds the centers of cluster_count clusters and groups the input samples around the clusters. As an output, \(\texttt{bestLabels}_i\) contains a 0based cluster index for the sample stored in the \(i^{th}\) row of the samples matrix.
data 
Data for clustering. An array of NDimensional points with float coordinates is needed. Examples of this array can be:

K  Number of clusters to split the set by. 
bestLabels  Input/output integer array that stores the cluster indices for every sample. 
criteria  The algorithm termination criteria, that is, the maximum number of iterations and/or the desired accuracy. The accuracy is specified as criteria.epsilon. As soon as each of the cluster centers moves by less than criteria.epsilon on some iteration, the algorithm stops. 
attempts  Flag to specify the number of times the algorithm is executed using different initial labellings. The algorithm returns the labels that yield the best compactness (see the last function parameter). 
flags  Flag that can take values of cv::KmeansFlags 
centers  Output matrix of the cluster centers, one row per each cluster center. 
\[\sum _i \ \texttt{samples} _i  \texttt{centers} _{ \texttt{labels} _i} \ ^2\]
after every attempt. The best (minimum) value is chosen and the corresponding labels and the compactness value are returned by the function. Basically, you can use only the core of the function, set the number of attempts to 1, initialize labels each time using a custom algorithm, pass them with the ( flags = KMEANS_USE_INITIAL_LABELS ) flag, and then choose the best (mostcompact) clustering.int partition  (  const std::vector< _Tp > &  _vec , 
std::vector< int > &  labels ,  
_EqPredicate 
predicate
=
_EqPredicate()


) 
#include <
opencv2/core/operations.hpp
>
Splits an element set into equivalency classes.
The generic function partition implements an \(O(N^2)\) algorithm for splitting a set of \(N\) elements into one or more equivalency classes, as described in http://en.wikipedia.org/wiki/Disjointset_data_structure . The function returns the number of equivalency classes.
_vec  Set of elements stored as a vector. 
labels 
Output vector of labels. It contains as many elements as vec. Each label labels[i] is a 0based cluster index of
vec[i]
.

predicate  Equivalence predicate (pointer to a boolean function of two arguments or an instance of the class that has the method bool operator()(const _Tp& a, const _Tp& b) ). The predicate returns true when the elements are certainly in the same class, and returns false if they may or may not be in the same class. 