opencv/samples/cpp/points_classifier.cpp

#include "opencv2/opencv_modules.hpp"
#include "opencv2/core/core.hpp"
#include "opencv2/ml/ml.hpp"
#include "opencv2/highgui/highgui.hpp"
#ifdef HAVE_OPENCV_OCL
#define _OCL_KNN_ 1 // select whether using ocl::KNN method or not, default is using
#define _OCL_SVM_ 1 // select whether using ocl::svm method or not, default is using
#include "opencv2/ocl/ocl.hpp"
#endif

#include <stdio.h>

using namespace std;
using namespace cv;
using namespace cv::ml;

const Scalar WHITE_COLOR = Scalar(255,255,255);
const string winName = "points";
const int testStep = 5;

Mat img, imgDst;
RNG rng;

vector<Point>  trainedPoints;
vector<int>    trainedPointsMarkers;
const int MAX_CLASSES = 2;
vector<Vec3b>  classColors(MAX_CLASSES);
int currentClass = 0;
vector<int> classCounters(MAX_CLASSES);

#define _NBC_ 1 // normal Bayessian classifier
#define _KNN_ 1 // k nearest neighbors classifier
#define _SVM_ 1 // support vectors machine
#define _DT_  1 // decision tree
#define _BT_  1 // ADA Boost
#define _GBT_ 0 // gradient boosted trees
#define _RF_  1 // random forest
#define _ANN_ 1 // artificial neural networks
#define _EM_  1 // expectation-maximization

static void on_mouse( int event, int x, int y, int /*flags*/, void* )
{
    if( img.empty() )
        return;

    int updateFlag = 0;

    if( event == EVENT_LBUTTONUP )
    {
        trainedPoints.push_back( Point(x,y) );
        trainedPointsMarkers.push_back( currentClass );
        classCounters[currentClass]++;
        updateFlag = true;
    }

    //draw
    if( updateFlag )
    {
        img = Scalar::all(0);

        // draw points
        for( size_t i = 0; i < trainedPoints.size(); i++ )
        {
            Vec3b c = classColors[trainedPointsMarkers[i]];
            circle( img, trainedPoints[i], 5, Scalar(c), -1 );
        }

        imshow( winName, img );
   }
}

static Mat prepare_train_samples(const vector<Point>& pts)
{
    Mat samples;
    Mat(pts).reshape(1, (int)pts.size()).convertTo(samples, CV_32F);
    return samples;
}

static Ptr<TrainData> prepare_train_data()
{
    Mat samples = prepare_train_samples(trainedPoints);
    return TrainData::create(samples, ROW_SAMPLE, Mat(trainedPointsMarkers));
}

static void predict_and_paint(const Ptr<StatModel>& model, Mat& dst)
{
    Mat testSample( 1, 2, CV_32FC1 );
    for( int y = 0; y < img.rows; y += testStep )
    {
        for( int x = 0; x < img.cols; x += testStep )
        {
            testSample.at<float>(0) = (float)x;
            testSample.at<float>(1) = (float)y;

            int response = (int)model->predict( testSample );
            dst.at<Vec3b>(y, x) = classColors[response];
        }
    }
}

#if _NBC_
static void find_decision_boundary_NBC()
{
    // learn classifier
    Ptr<NormalBayesClassifier> normalBayesClassifier = NormalBayesClassifier::create();
    normalBayesClassifier->train(prepare_train_data());

    predict_and_paint(normalBayesClassifier, imgDst);
}
#endif


#if _KNN_
static void find_decision_boundary_KNN( int K )
{
    Ptr<KNearest> knn = KNearest::create(true);
    knn->setDefaultK(K);
    knn->train(prepare_train_data());

    predict_and_paint(knn, imgDst);
}
#endif

#if _SVM_
static void find_decision_boundary_SVM( SVM::Params params )
{
    Ptr<SVM> svm = SVM::create(params);
    svm->train(prepare_train_data());

    predict_and_paint(svm, imgDst);

    Mat sv = svm->getSupportVectors();
    for( int i = 0; i < sv.rows; i++ )
    {
        const float* supportVector = sv.ptr<float>(i);
        circle( imgDst, Point(saturate_cast<int>(supportVector[0]),saturate_cast<int>(supportVector[1])), 5, Scalar(255,255,255), -1 );
    }
}
#endif

#if _DT_
static void find_decision_boundary_DT()
{
    DTrees::Params params;
    params.maxDepth = 8;
    params.minSampleCount = 2;
    params.useSurrogates = false;
    params.CVFolds = 0; // the number of cross-validation folds
    params.use1SERule = false;
    params.truncatePrunedTree = false;

    Ptr<DTrees> dtree = DTrees::create(params);
    dtree->train(prepare_train_data());

    predict_and_paint(dtree, imgDst);
}
#endif

#if _BT_
static void find_decision_boundary_BT()
{
    Boost::Params params( Boost::DISCRETE, // boost_type
                          100, // weak_count
                          0.95, // weight_trim_rate
                          2, // max_depth
                          false, //use_surrogates
                          Mat() // priors
                          );

    Ptr<Boost> boost = Boost::create(params);
    boost->train(prepare_train_data());
    predict_and_paint(boost, imgDst);
}

#endif

#if _GBT_
static void find_decision_boundary_GBT()
{
    GBTrees::Params params( GBTrees::DEVIANCE_LOSS, // loss_function_type
                         100, // weak_count
                         0.1f, // shrinkage
                         1.0f, // subsample_portion
                         2, // max_depth
                         false // use_surrogates )
                         );

    Ptr<GBTrees> gbtrees = GBTrees::create(params);
    gbtrees->train(prepare_train_data());
    predict_and_paint(gbtrees, imgDst);
}
#endif

#if _RF_
static void find_decision_boundary_RF()
{
    RTrees::Params  params( 4, // max_depth,
                        2, // min_sample_count,
                        0.f, // regression_accuracy,
                        false, // use_surrogates,
                        16, // max_categories,
                        Mat(), // priors,
                        false, // calc_var_importance,
                        1, // nactive_vars,
                        TermCriteria(TermCriteria::MAX_ITER, 5, 0) // max_num_of_trees_in_the_forest,
                       );

    Ptr<RTrees> rtrees = RTrees::create(params);
    rtrees->train(prepare_train_data());
    predict_and_paint(rtrees, imgDst);
}

#endif

#if _ANN_
static void find_decision_boundary_ANN( const Mat&  layer_sizes )
{
    ANN_MLP::Params params(TermCriteria(TermCriteria::MAX_ITER+TermCriteria::EPS, 300, FLT_EPSILON),
                           ANN_MLP::Params::BACKPROP, 0.001);
    Ptr<ANN_MLP> ann = ANN_MLP::create(layer_sizes, params, ANN_MLP::SIGMOID_SYM, 1, 1 );

    Mat trainClasses = Mat::zeros( trainedPoints.size(), classColors.size(), CV_32FC1 );
    for( int i = 0; i < trainClasses.rows; i++ )
    {
        trainClasses.at<float>(i, trainedPointsMarkers[i]) = 1.f;
    }

    Mat samples = prepare_train_samples(trainedPoints);
    Ptr<TrainData> tdata = TrainData::create(samples, ROW_SAMPLE, trainClasses);

    ann->train(tdata);

    Mat testSample( 1, 2, CV_32FC1 );
    Mat outputs;
    for( int y = 0; y < img.rows; y += testStep )
    {
        for( int x = 0; x < img.cols; x += testStep )
        {
            testSample.at<float>(0) = (float)x;
            testSample.at<float>(1) = (float)y;

            ann->predict( testSample, outputs );
            Point maxLoc;
            minMaxLoc( outputs, 0, 0, 0, &maxLoc );
            imgDst.at<Vec3b>(y, x) = classColors[maxLoc.x];
        }
    }
}
#endif

#if _EM_
static void find_decision_boundary_EM()
{
    img.copyTo( imgDst );

    Mat samples = prepare_train_samples(trainedPoints);

    int i, j, nmodels = (int)classColors.size();
    vector<Ptr<EM> > em_models(nmodels);
    Mat modelSamples;

    for( i = 0; i < nmodels; i++ )
    {
        const int componentCount = 3;

        modelSamples.release();
        for( j = 0; j < samples.rows; j++ )
        {
            if( trainedPointsMarkers[j] == i )
                modelSamples.push_back(samples.row(j));
        }

        // learn models
        if( !modelSamples.empty() )
        {
            em_models[i] = EM::train(modelSamples, noArray(), noArray(), noArray(),
                                   EM::Params(componentCount, EM::COV_MAT_DIAGONAL));
        }
    }

    // classify coordinate plane points using the bayes classifier, i.e.
    // y(x) = arg max_i=1_modelsCount likelihoods_i(x)
    Mat testSample(1, 2, CV_32FC1 );
    Mat logLikelihoods(1, nmodels, CV_64FC1, Scalar(-DBL_MAX));

    for( int y = 0; y < img.rows; y += testStep )
    {
        for( int x = 0; x < img.cols; x += testStep )
        {
            testSample.at<float>(0) = (float)x;
            testSample.at<float>(1) = (float)y;

            for( i = 0; i < nmodels; i++ )
            {
                if( !em_models[i].empty() )
                    logLikelihoods.at<double>(i) = em_models[i]->predict2(testSample, noArray())[0];
            }
            Point maxLoc;
            minMaxLoc(logLikelihoods, 0, 0, 0, &maxLoc);
            imgDst.at<Vec3b>(y, x) = classColors[maxLoc.x];
        }
    }
}
#endif

int main()
{
    cout << "Use:" << endl
         << "  key '0' .. '1' - switch to class #n" << endl
         << "  left mouse button - to add new point;" << endl
         << "  key 'r' - to run the ML model;" << endl
         << "  key 'i' - to init (clear) the data." << endl << endl;

    cv::namedWindow( "points", 1 );
    img.create( 480, 640, CV_8UC3 );
    imgDst.create( 480, 640, CV_8UC3 );

    imshow( "points", img );
    setMouseCallback( "points", on_mouse );

    classColors[0] = Vec3b(0, 255, 0);
    classColors[1] = Vec3b(0, 0, 255);

    for(;;)
    {
        uchar key = (uchar)waitKey();

        if( key == 27 ) break;

        if( key == 'i' ) // init
        {
            img = Scalar::all(0);

            trainedPoints.clear();
            trainedPointsMarkers.clear();
            classCounters.assign(MAX_CLASSES, 0);

            imshow( winName, img );
        }

        if( key == '0' || key == '1' )
        {
            currentClass = key - '0';
        }

        if( key == 'r' ) // run
        {
            double minVal = 0;
            minMaxLoc(classCounters, &minVal, 0, 0, 0);
            if( minVal == 0 )
            {
                printf("each class should have at least 1 point\n");
                continue;
            }
            img.copyTo( imgDst );
#if _NBC_
            find_decision_boundary_NBC();
            namedWindow( "NormalBayesClassifier", WINDOW_AUTOSIZE );
            imshow( "NormalBayesClassifier", imgDst );
#endif
#if _KNN_
            int K = 3;
            find_decision_boundary_KNN( K );
            namedWindow( "kNN", WINDOW_AUTOSIZE );
            imshow( "kNN", imgDst );

            K = 15;
            find_decision_boundary_KNN( K );
            namedWindow( "kNN2", WINDOW_AUTOSIZE );
            imshow( "kNN2", imgDst );
#endif

#if _SVM_
            //(1)-(2)separable and not sets
            SVM::Params params;
            params.svmType = SVM::C_SVC;
            params.kernelType = SVM::POLY; //CvSVM::LINEAR;
            params.degree = 0.5;
            params.gamma = 1;
            params.coef0 = 1;
            params.C = 1;
            params.nu = 0.5;
            params.p = 0;
            params.termCrit = TermCriteria(TermCriteria::MAX_ITER+TermCriteria::EPS, 1000, 0.01);

            find_decision_boundary_SVM( params );
            namedWindow( "classificationSVM1", WINDOW_AUTOSIZE );
            imshow( "classificationSVM1", imgDst );

            params.C = 10;
            find_decision_boundary_SVM( params );
            namedWindow( "classificationSVM2", WINDOW_AUTOSIZE );
            imshow( "classificationSVM2", imgDst );
#endif

#if _DT_
            find_decision_boundary_DT();
            namedWindow( "DT", WINDOW_AUTOSIZE );
            imshow( "DT", imgDst );
#endif

#if _BT_
            find_decision_boundary_BT();
            namedWindow( "BT", WINDOW_AUTOSIZE );
            imshow( "BT", imgDst);
#endif

#if _GBT_
            find_decision_boundary_GBT();
            namedWindow( "GBT", WINDOW_AUTOSIZE );
            imshow( "GBT", imgDst);
#endif

#if _RF_
            find_decision_boundary_RF();
            namedWindow( "RF", WINDOW_AUTOSIZE );
            imshow( "RF", imgDst);
#endif

#if _ANN_
            Mat layer_sizes1( 1, 3, CV_32SC1 );
            layer_sizes1.at<int>(0) = 2;
            layer_sizes1.at<int>(1) = 5;
            layer_sizes1.at<int>(2) = classColors.size();
            find_decision_boundary_ANN( layer_sizes1 );
            namedWindow( "ANN", WINDOW_AUTOSIZE );
            imshow( "ANN", imgDst );
#endif

#if _EM_
            find_decision_boundary_EM();
            namedWindow( "EM", WINDOW_AUTOSIZE );
            imshow( "EM", imgDst );
#endif
        }
    }

    return 1;
}