K Nearest Neighbors in CSharp

This example requires Emgu CV 1.5.0.0 or later release

What is a K Nearest Neighbors Classifier

According to wikipedia,

In pattern recognition, the k-nearest neighbors algorithm (k-NN) is a method for classifying objects based on closest training examples in the feature space. k-NN is a type of instance-based learning, or lazy learning where the function is only approximated locally and all computation is deferred until classification. It can also be used for regression.

Source Code

Emgu CV 3.x

Click to view source code

using System.Drawing;
using Emgu.CV.Structure;
using Emgu.CV.ML;
using Emgu.CV.ML.Structure;

...


int K = 10;
int trainSampleCount = 100;

#region Generate the training data and classes

Matrix<float> trainData = new Matrix<float>(trainSampleCount, 2);
Matrix<float> trainClasses = new Matrix<float>(trainSampleCount, 1);

Image<Bgr, Byte> img = new Image<Bgr, byte>(500, 500);

Matrix<float> sample = new Matrix<float>(1, 2);

Matrix<float> trainData1 = trainData.GetRows(0, trainSampleCount >> 1, 1);
trainData1.SetRandNormal(new MCvScalar(200), new MCvScalar(50));
Matrix<float> trainData2 = trainData.GetRows(trainSampleCount >> 1, trainSampleCount, 1);
trainData2.SetRandNormal(new MCvScalar(300), new MCvScalar(50));

Matrix<float> trainClasses1 = trainClasses.GetRows(0, trainSampleCount >> 1, 1);
trainClasses1.SetValue(1);
Matrix<float> trainClasses2 = trainClasses.GetRows(trainSampleCount >> 1, trainSampleCount, 1);
trainClasses2.SetValue(2);
#endregion
         
Matrix<float> results, neighborResponses;
results = new Matrix<float>(sample.Rows, 1);
neighborResponses = new Matrix<float>(sample.Rows, K);
//dist = new Matrix<float>(sample.Rows, K);

using (KNearest knn = new KNearest())
{
   knn.DefaultK = K;
   knn.IsClassifier = true;
   knn.Train(trainData, MlEnum.DataLayoutType.RowSample, trainClasses);

   for (int i = 0; i < img.Height; i++)
   {
      for (int j = 0; j < img.Width; j++)
      {
         sample.Data[0, 0] = j;
         sample.Data[0, 1] = i;

         // estimates the response and get the neighbors' labels
         float response = knn.Predict(sample); //knn.FindNearest(sample, K, results, null, neighborResponses, null);

         int accuracy = 0;
         // compute the number of neighbors representing the majority
         for (int k = 0; k < K; k++)
         {
            if (neighborResponses.Data[0, k] == response)
               accuracy++;
         }
         // highlight the pixel depending on the accuracy (or confidence)
         img[i, j] = 
            response == 1 ?
               (accuracy > 5 ? new Bgr(90, 0, 0) : new Bgr(90, 40, 0)) :
               (accuracy > 5 ? new Bgr(0, 90, 0) : new Bgr(40, 90, 0));
      }
   }

}

// display the original training samples
for (int i = 0; i < (trainSampleCount >> 1); i++)
{
   PointF p1 = new PointF(trainData1[i, 0], trainData1[i, 1]);
   img.Draw(new CircleF(p1, 2.0f), new Bgr(255, 100, 100), -1);
   PointF p2 = new PointF(trainData2[i, 0], trainData2[i, 1]);
   img.Draw(new CircleF(p2, 2.0f), new Bgr(100, 255, 100), -1);
}

Emgu.CV.UI.ImageViewer.Show(img);

Emgu CV 2.x

Click to view source code

using System.Drawing;
using Emgu.CV.Structure;
using Emgu.CV.ML;
using Emgu.CV.ML.Structure;

...

int K = 10;
int trainSampleCount = 100;

#region Generate the traning data and classes

Matrix<float> trainData = new Matrix<float>(trainSampleCount, 2);
Matrix<float> trainClasses = new Matrix<float>(trainSampleCount, 1);

Image<Bgr, Byte> img = new Image<Bgr, byte>(500, 500);

Matrix<float> sample = new Matrix<float>(1, 2);

Matrix<float> trainData1 = trainData.GetRows(0, trainSampleCount >> 1, 1);
trainData1.SetRandNormal(new MCvScalar(200), new MCvScalar(50));
Matrix<float> trainData2 = trainData.GetRows(trainSampleCount >> 1, trainSampleCount, 1);
trainData2.SetRandNormal(new MCvScalar(300), new MCvScalar(50));

Matrix<float> trainClasses1 = trainClasses.GetRows(0, trainSampleCount >> 1, 1);
trainClasses1.SetValue(1);
Matrix<float> trainClasses2 = trainClasses.GetRows(trainSampleCount >> 1, trainSampleCount, 1);
trainClasses2.SetValue(2);
#endregion

Matrix<float> results, neighborResponses;
results = new Matrix<float>(sample.Rows, 1);
neighborResponses = new Matrix<float>(sample.Rows, K);
//dist = new Matrix<float>(sample.Rows, K);

using (KNearest knn = new KNearest(trainData, trainClasses, null, false, K))
{
   for (int i = 0; i < img.Height; i++)
   {
      for (int j = 0; j < img.Width; j++)
      {
         sample.Data[0, 0] = j;
         sample.Data[0, 1] = i;

         //Matrix<float> nearestNeighbors = new Matrix<float>(K* sample.Rows, sample.Cols);
         // estimates the response and get the neighbors' labels
         float response = knn.FindNearest(sample, K, results, null, neighborResponses, null);

         int accuracy = 0;
         // compute the number of neighbors representing the majority
         for (int k = 0; k < K; k++)
         {
            if (neighborResponses.Data[0, k] == response)
               accuracy++;
         }
         // highlight the pixel depending on the accuracy (or confidence)
         img[i, j] =
         response == 1 ?
             (accuracy > 5 ? new Bgr(90, 0, 0) : new Bgr(90, 60, 0)) :
             (accuracy > 5 ? new Bgr(0, 90, 0) : new Bgr(60, 90, 0));
      }
   }
}

// display the original training samples
for (int i = 0; i < (trainSampleCount >> 1); i++)
{
   PointF p1 = new PointF(trainData1[i, 0], trainData1[i, 1]);
   img.Draw(new CircleF(p1, 2.0f), new Bgr(255, 100, 100), -1);
   PointF p2 = new PointF(trainData2[i, 0], trainData2[i, 1]);
   img.Draw(new CircleF(p2, 2.0f), new Bgr(100, 255, 100), -1);
}

Emgu.CV.UI.ImageViewer.Show(img);

Result