1 #include "opencv2/highgui.hpp"
2 #include "opencv2/ml.hpp"
5 using namespace cv::ml;
7 int main( int /*argc*/, char** /*argv*/ )
10 const int N1 = (int)sqrt((double)N);
11 const Scalar colors[] =
13 Scalar(0,0,255), Scalar(0,255,0),
14 Scalar(0,255,255),Scalar(255,255,0)
19 Mat samples( nsamples, 2, CV_32FC1 );
21 Mat img = Mat::zeros( Size( 500, 500 ), CV_8UC3 );
22 Mat sample( 1, 2, CV_32FC1 );
24 samples = samples.reshape(2, 0);
25 for( i = 0; i < N; i++ )
27 // form the training samples
28 Mat samples_part = samples.rowRange(i*nsamples/N, (i+1)*nsamples/N );
30 Scalar mean(((i%N1)+1)*img.rows/(N1+1),
31 ((i/N1)+1)*img.rows/(N1+1));
33 randn( samples_part, mean, sigma );
35 samples = samples.reshape(1, 0);
38 Ptr<EM> em_model = EM::train( samples, noArray(), labels, noArray(),
39 EM::Params(N, EM::COV_MAT_SPHERICAL,
40 TermCriteria(TermCriteria::COUNT+TermCriteria::EPS, 300, 0.1)));
42 // classify every image pixel
43 for( i = 0; i < img.rows; i++ )
45 for( j = 0; j < img.cols; j++ )
47 sample.at<float>(0) = (float)j;
48 sample.at<float>(1) = (float)i;
49 int response = cvRound(em_model->predict2( sample, noArray() )[1]);
50 Scalar c = colors[response];
52 circle( img, Point(j, i), 1, c*0.75, FILLED );
56 //draw the clustered samples
57 for( i = 0; i < nsamples; i++ )
59 Point pt(cvRound(samples.at<float>(i, 0)), cvRound(samples.at<float>(i, 1)));
60 circle( img, pt, 1, colors[labels.at<int>(i)], FILLED );
63 imshow( "EM-clustering result", img );