C++でOpenCVを使ってK-Meansクラスタリングを行うサンプル。

OpenCVのインストール

CentOSの場合

yum install -y opencv-devel

サンプルソース

#include <iostream>

#include <opencv2/core/core.hpp>
#include <opencv2/flann/flann.hpp>

int main() {
  cv::Mat samples = cv::Mat::zeros(5, 5, CV_32FC1);
  samples.at<float>(0, 0) = 100.0;
  samples.at<float>(1, 0) = 100.0;
  samples.at<float>(2, 0) = 80.0;
  samples.at<float>(3, 0) = 70.0;
  samples.at<float>(4, 0) = 50.0;
  std::cout << "samples=\n" << samples << std::endl;

  int n_clusters = 3;
  cv::Mat clusters = cv::Mat::zeros(samples.rows, 1, CV_32SC1);

  cv::Mat centers;
  cv::theRNG() = 19771228; //クラスタリングの結果を固定する場合
  cv::kmeans(samples,
             n_clusters,
             clusters,
             cvTermCriteria(CV_TERMCRIT_EPS|CV_TERMCRIT_ITER, 10, 1.0),
             1,
             cv::KMEANS_PP_CENTERS,
             centers);

  //std::cout << clusters << std::endl;
  printf("clusters=\n");
  for (int i = 0; i < clusters.rows; i++) {
    printf("%d ", clusters.at<int>(i));
  }
  printf("\n");

  //std::cout << centers << std::endl;
  printf("centers=\n");
  for (int i = 0; i < centers.rows; i++) {
    for (int j = 0; j < centers.cols; j++) {
      printf("%f ", centers.at<float>(i, j));
    }
    printf("\n");
  }
  return 0;
}

コンパイル

g++ -I /usr/include/opencv2 -lopencv_core -lopencv_flann kmeans.cpp -o kmeans

実行結果

 ./kmeans
samples=
[100, 0, 0, 0, 0;
  100, 0, 0, 0, 0;
  80, 0, 0, 0, 0;
  70, 0, 0, 0, 0;
  50, 0, 0, 0, 0]
clusters=
0 0 1 1 2
centers=
100.000000 0.000000 0.000000 0.000000 0.000000
75.000000 0.000000 0.000000 0.000000 0.000000
50.000000 0.000000 0.000000 0.000000 0.000000

参考