public void ShowClusters(int[,] img, ClustersGenerator.ClusterPoint[] centroids, int W)
{
Normalizer.Normalize(100, 500, img);
int maxY = img.GetLength(0) / W;
int maxX = img.GetLength(1) / W;
var p = new Pen(Color.White);
var image = ImageHelper.SaveArrayToBitmap(img.Select2D(x => (double)x));
var g = Graphics.FromImage(image);
var info = RidgeFrequencyGenerator.GenerateBlocksInfo(img);
for (int i = 0; i < maxY; i++)
{
for (int j = 0; j < maxX; j++)
{
g.DrawLine(p, new Point(j * W, i * W), new Point(j * W, i * W + W));
g.DrawLine(p, new Point(j * W, i * W), new Point(j * W + W, i * W));
g.DrawLine(p, new Point(j * W + W, i * W + W), new Point(j * W, i * W + W));
g.DrawLine(p, new Point(j * W + W, i * W + W), new Point(j * W + W, i * W));
int cl = info[i * W + j].FindClosestIndex(centroids);
g.DrawString(cl.ToString(), new Font("Times New Roman", 12), Brushes.Black, new PointF(j * W, i * W));
}
}
g.Save();
var path = Path.GetTempPath() + "1.png";
var tmp = ImageHelper.LoadImageAsInt(image);
ImageHelper.SaveIntArray(tmp, path);
Process.Start(path);
}
public void TestClusterization()
{
const int W = 16;
var img = ImageHelper.LoadImageAsInt(TestResources._1);
Normalizer.Normalize(100, 500, img);
var info = RidgeFrequencyGenerator.GenerateBlocksInfo(img);
var centers = new ClustersGenerator.ClusterPoint[6];
img = ImageHelper.LoadImageAsInt(TestResources._1);
Normalizer.Normalize(100, 500, img);
info.AddRange(RidgeFrequencyGenerator.GenerateBlocksInfo(img));
img = ImageHelper.LoadImageAsInt(TestResources._2);
Normalizer.Normalize(100, 500, img);
info.AddRange(RidgeFrequencyGenerator.GenerateBlocksInfo(img));
img = ImageHelper.LoadImageAsInt(TestResources._3);
Normalizer.Normalize(100, 500, img);
info.AddRange(RidgeFrequencyGenerator.GenerateBlocksInfo(img));
img = ImageHelper.LoadImageAsInt(TestResources._4);
Normalizer.Normalize(100, 500, img);
info.AddRange(RidgeFrequencyGenerator.GenerateBlocksInfo(img));
img = ImageHelper.LoadImageAsInt(TestResources._5);
Normalizer.Normalize(100, 500, img);
var middle = new ClustersGenerator.ClusterPoint(0, 0, 0);
foreach (var bl in info)
{
middle.Amplitude += bl.Amplitude;
middle.Frequency += bl.Frequency;
middle.Variance += bl.Variance;
}
middle.Amplitude /= info.Count;
middle.Frequency /= info.Count;
middle.Variance /= info.Count;
Random rand = new Random();
for (int i = 0; i < 6; i++)
{
centers[i] = info[rand.Next(0, info.Count)];
}
var centroids = ClustersGenerator.Clusterization(info, centers);
// Writing result to file
var path = ClustersGenerator.ClusterPath;
using (StreamWriter file = new StreamWriter(path))
{
for (int i = 0; i < centers.Length; i++)
{
file.WriteLine(i);
file.WriteLine(centroids[i].Amplitude);
file.WriteLine(centroids[i].Frequency);
file.WriteLine(centroids[i].Variance);
file.WriteLine();
}
}
Process.Start(path);
img = ImageHelper.LoadImageAsInt(TestResources._1);
ShowClusters(img, centroids, W);
img = ImageHelper.LoadImageAsInt(TestResources._2);
ShowClusters(img, centroids, W);
img = ImageHelper.LoadImageAsInt(TestResources._3);
ShowClusters(img, centroids, W);
img = ImageHelper.LoadImageAsInt(TestResources._4);
ShowClusters(img, centroids, W);
img = ImageHelper.LoadImageAsInt(TestResources._5);
ShowClusters(img, centroids, W);
}