Revision history [back]

(Include header files)
using namespace cv;
using namespace cv::ml;
using namespace std;
void readCenters(cv::Mat&, const char *);
 
*);
int CLUSTER_COUNT=5;//number of clusters
Mat train_data;
Mat test_data;
 
test_data;
void printDim(cv::Mat mat,const char *name)
 
*name)
void splitData(cv::Mat &data,cv::Mat &train_data,cv::Mat &test_data)
 
&test_data)
void readData(cv::Mat& data)
{
std::vector<const char="" *=""> char *> filenames;
filenames.push_back("Data/ik147_1.txt");
filenames.push_back("Data/ik147_2.txt");
string line;
vector<mat> vector<Mat> raw_data(2);//= new std::vector<cv::mat>(4);
std::vector<cv::Mat>(4);
 int row;
double min,max;
for(int i =0;i<2;i++)
{
ifstream file( filenames[i] );
while( file>>row )
{
raw_data[i].push_back(row);
}
minMaxLoc(raw_data[i],&min,&max);
cout<<filenames[i]&lt;&lt;" min="" :"&lt;<min&lt;&lt;",="" max="" :"&lt;<max&lt;<std::endl;<="" p="">
cout<<filenames[i]<<" min :"<<min<<", max :"<<max<<std::endl;
}
 int N=raw_data[0].rows;
 
 // cv::Mat data(N,3,CV_32FC1);
cout<<"N value : "<< N << endl;
int columns_to_read=2;
data.create(N,columns_to_read,CV_32FC1);
 
data.create(N,columns_to_read,CV_32FC1);
for(int i=0;i<columns_to_read;i++)
 {
 raw_data[i](Rect(0,0,1,N)).copyTo(data(Rect(i,0,1,N)));
 }
 
}
 

}
void computeLabelledData(Mat& data,Mat &data_with_labels)
{
Mat labels,centers; 
 
kmeans(data, CLUSTER_COUNT, labels,
TermCriteria( CV_TERMCRIT_EPS+CV_TERMCRIT_ITER, 10, 1.0),
3, KMEANS_PP_CENTERS, centers);
 
centers);
data_with_labels.create(data.rows,data.cols+1,CV_32FC1);
 data.copyTo(data_with_labels(Rect(0,0,data.cols,data.rows)));
 labels.copyTo(data_with_labels(Rect(data.cols,0,labels.cols,labels.rows)));
 
}
 
}
void print(Mat mat, int prec)
{ 
 
for(int i=0; i<mat.size().height; i++)="" {="" cout="" &lt;&lt;="" "[";="" for(int="" j="0;" j<mat.size().width;="" j++)="" {="" cout="" &lt;&lt;="" setprecision(prec)="" &lt;&lt;="" mat.at<float="">(i,j);
i++)
{
cout << "[";
for(int j=0; j<mat.size().width; j++)
{
cout << setprecision(prec) << mat.at<float>(i,j);
 if(j != mat.size().width-1)
cout << ", ";
else
cout << "]" << endl;
}
}
}
int main()
{
// Data for visual representation
int width = 512, height = 512;
Mat image = Mat::zeros(height, width, CV_8UC3);
 
CV_8UC3);
 Mat data;
 readData(data);
 printDim(data,"data");
 Mat data_with_labels,train_data_labels,test_data_labels;
 computeLabelledData(data,data_with_labels);
 splitData(data_with_labels,train_data_labels,test_data_labels);
 //splitData(data,train_data,test_data);
 Mat train_data,train_labels;
 train_data_labels(cv::Rect(0,0,train_data_labels.cols-1,train_data_labels.rows)).copyTo(train_data);
 train_data_labels(cv::Rect(train_data_labels.cols-1,0,1,train_data_labels.rows)).copyTo(train_labels);
 Mat test_data,test_labels,predicted_labels,confusion_matrix;
 test_data_labels(cv::Rect(0,0,test_data_labels.cols-1,test_data_labels.rows)).copyTo(test_data);
 test_data_labels(cv::Rect(test_data_labels.cols-1,0,1,test_data_labels.rows)).copyTo(test_labels);
 printDim(train_data,"train_data");
 printDim(test_data,"test_data");
 cout <<"train data : " << endl;
 //print(train_data, 3);
 cout <<"test data : " << endl;
 //print(test_data, 3);
 int N=test_data.rows; //number of data points
 int K=5; //number of labels
 RNG rng((unsigned)time(NULL));
 RNG rng_center(20);
 int clusterCount=K;
 int sampleCount = N;
 Mat centers(5,2,CV_32FC1);
 readCenters(centers,"centers.txt");
 Mat labelsMat(N,1,CV_32SC1);
 Mat labelsMatKMeans;
 Mat img_kmeans(512, 512, CV_8UC3);
 Point center;
 center.x=0;//centers.at<float>(k,0);
 center.y=0;//:centers.at<float>(k,1);
 // Set up training data
 /* generate random labeld sample from multigaussian distribution */
 for( int k = 0; k < clusterCount; ++k )
 {
  Mat pointChunk = test_data.rowRange(k*sampleCount/clusterCount,
  k == clusterCount - 1 ? sampleCount :
 (k+1)*sampleCount/clusterCount);
  rng.fill(pointChunk, CV_RAND_NORMAL, Scalar(center.x, center.y,0,0), Scalar(width*0.10, height*0.10));
 Mat repeat_center;
 repeat(centers.row(k),pointChunk.rows,1,repeat_center);
 pointChunk=pointChunk+repeat_center;
  Mat labelChunk = labelsMat.rowRange(k*sampleCount/clusterCount,
  k == clusterCount - 1 ? sampleCount :
 (k+1)*sampleCount/clusterCount);
 labelChunk=k;
 }
 test_data=abs(test_data);
 for(int i=0;i<N;++i)
 {
  test_data.at<float>(i,0)= cvCeil(test_data.at<float>(i,0));
 test_data.at<float>(i,1)= cvCeil(test_data.at<float>(i,1));
 }
  std::vector<Vec3b> colors_region;
 std::vector<Vec3b> colors_data;
 //generate random colors for each label
 for(int i=0;i<K;++i)
 {
  int icolor = (unsigned) rng;
  colors_region.push_back( Vec3b( icolor&255, (icolor>>8)&255, (icolor>>16)&255 ));
 icolor=(unsigned)rng;
  colors_data.push_back( Vec3b( icolor&255, (icolor>>8)&255, (icolor>>16)&255 ));
 }
 
 //draw k-means output
kmeans(test_data, clusterCount, labelsMatKMeans,
TermCriteria( CV_TERMCRIT_EPS+CV_TERMCRIT_ITER, 10, 1.0),
3, KMEANS_PP_CENTERS, centers);
 
centers);
 print(centers, 3);
  img_kmeans = Scalar::all(0);
  for( int i = 0; i < sampleCount; i++ )
 {
  int clusterIdx = labelsMatKMeans.at<int>(i);
 Point ipt ;
 ipt.x= (int)test_data.at<float>(i,0);
 ipt.y=(int)test_data.at<float>(i,1);
  circle( img_kmeans, ipt, 5, Scalar(colors_data[clusterIdx]), -1, 8 );
 }
  imshow("K-Means clustering", img_kmeans);
 imwrite("k_means.png",img_kmeans);
 
// Ptr<svm> Ptr<SVM> svm = Algorithm::load<svm>("svm_params.xml");
 
Algorithm::load<SVM>("svm_params.xml");
cout<<"data loaded"<< endl;
 waitKey(0);
 
}
 

}
void readCenters(cv::Mat &centers, const char * filename){
const int ROWS=5;
const int COLS=2;
std::cout<<"reading centers "<<filename&lt;<std::endl; float="" array[rows][cols];="" std::ifstream="" file(="" filename="" );="" std::string="" line;="" int="" row,col;="" int="" i="0;&lt;/p">
"<<filename<<std::endl;
float array[ROWS][COLS];
std::ifstream file( filename );
std::string line;
int row,col;
int i=0;
while( file>>row>>col )
 {
 centers.at<float>(i,0)=row;
 centers.at<float>(i,1)=col;
 i++;
  cout<<"center "<< i <<": ("<< row <<","<< col << ")" <<endl;
 }
}
  
}