/// <summary>
/// Get the coefficient map of inner structures of tissue for the transform rigidity penalty term.
/// </summary>
/// <param name="filename">image filename</param>
/// <returns>return coefficient map filename</returns>
private string GetInnerStructureSegmentationsAsCoefficientMap(string filename)
{
InnerTissueSegmentation innerSegImage = GetInnerStructureSegmentation(filename);
string filenameCoefficientMap = ReadWriteUtils.GetOutputDirectory(_parameters, _parameters.Iteration) + Constants.cCoefficientFilename;
ReadWriteUtils.WriteUMatToFile(filenameCoefficientMap, innerSegImage.GetOutput().FirstOrDefault());
innerSegImage.Dispose();
// rescale image
sitk.Image img = ReadWriteUtils.ReadITKImageFromFile(filenameCoefficientMap);
sitk.CastImageFilter castFilter = new sitk.CastImageFilter();
castFilter.SetOutputPixelType(sitk.PixelIDValueEnum.sitkFloat32);
img = castFilter.Execute(img);
sitk.RescaleIntensityImageFilter filter = new sitk.RescaleIntensityImageFilter();
filter.SetOutputMinimum(0.0);
filter.SetOutputMaximum(1.0);
sitk.Image coefficientMap = filter.Execute(img);
// save as mhd
filenameCoefficientMap = ReadWriteUtils.GetOutputDirectory(_parameters, _parameters.Iteration) + Constants.cCoefficientFilename;
ReadWriteUtils.WriteSitkImage(coefficientMap, filenameCoefficientMap);
coefficientMap.Dispose();
return(filenameCoefficientMap);
}
private string DoWholeTissueSegmentation(Image <Bgr, byte> img, out Image <Gray, byte> mask, string filename)
{
WholeTissueSegmentation seg = new WholeTissueSegmentation(img, registrationParameters.WholeTissueSegParams);
seg.Execute();
mask = seg.GetOutput().Clone();
seg.Dispose();
string filepathMask = registrationParameters.OutputDirectory + filename;
ReadWriteUtils.WriteUMatToFile(filepathMask, mask.ToUMat());
return(filepathMask);
}
private string DoInnerTissueSegmentation(Image <Bgr, byte> img, Image <Gray, byte> mask, string filename)
{
InnerTissueSegmentation seg = new InnerTissueSegmentation(img.Clone(), mask.Clone(), registrationParameters.InnerStructuresSegParams);
seg.Execute();
UMat innerMask = seg.GetOutput()[0].Clone();
seg.Dispose();
string filepathResult = registrationParameters.OutputDirectory + filename;
ReadWriteUtils.WriteUMatToFile(filepathResult, innerMask);
return(filepathResult);
}
/// <summary>
/// Kmeans segmentation of given image.
/// 1. Do a kmeans clustering
/// 2. Find contours
/// 3. Classification of clusters (?)
/// </summary>
/// <param name="uMatChannel"></param>
private void KmeansSegmentation(UMat uMatChannel)
{
if (false)
{
Image <Bgr, byte> newImage = new Image <Bgr, byte>(uMatChannel.Bitmap);
Image <Bgr, byte> clusteredImage = SegmentationUtils.KMeansClustering <Bgr, byte, byte>(newImage.Clone(), 3, SegmentationUtils._clusterColors);
ReadWriteUtils.WriteUMatToFile(@"D:\testdata\_temp\kmeans.png", clusteredImage.ToUMat());
// TODO
// Find contours
// Classificate contours
}
throw new NotImplementedException();
}
/// <summary>
/// Generic method to apply k-Means-Clustering on an opencv image of unknown type.
/// </summary>
/// <typeparam name="T">color space type of the image</typeparam>
/// <typeparam name="V">pixel data type</typeparam>
/// <typeparam name="K">color / intensities datatype</typeparam>
/// <param name="image">input image</param>
/// <param name="k">amount of clusters k</param>
/// <param name="colors">array of colors used to represent the clusters</param>
/// <returns>clustered image</returns>
public static Image <T, V> KMeansClustering <T, V, K>(Image <T, K> image, int k, T[] colors) where T : struct, IColor where K : new() where V : new()
{
if (k > 4)
{
return(null);
}
Image <Bgr, float> src = image.Convert <Bgr, float>();
Matrix <float> samples = new Matrix <float>(src.Rows * src.Cols, 1, 3);
Matrix <int> finalClusters = new Matrix <int>(src.Rows * src.Cols, 1);
for (int y = 0; y < src.Rows; y++)
{
for (int x = 0; x < src.Cols; x++)
{
samples.Data[y + x * src.Rows, 0] = (float)src[y, x].Blue;
samples.Data[y + x * src.Rows, 1] = (float)src[y, x].Green;
samples.Data[y + x * src.Rows, 2] = (float)src[y, x].Red;
}
}
MCvTermCriteria criteria = new MCvTermCriteria(50, 0.5);
criteria.Type = TermCritType.Iter | TermCritType.Eps;
Matrix <Single> centers = new Matrix <Single>(k, src.Rows * src.Cols);
double compactness = CvInvoke.Kmeans(samples, k, finalClusters, criteria, 3, KMeansInitType.RandomCenters);
Image <T, V> kMeansImage = new Image <T, V>(src.Size);
for (int y = 0; y < src.Rows; y++)
{
for (int x = 0; x < src.Cols; x++)
{
PointF p = new PointF(x, y);
kMeansImage.Draw(new CircleF(p, 1.0f), colors[finalClusters[y + x * src.Rows, 0]], 1);
}
}
// Debug
ReadWriteUtils.WriteUMatToFile(ApplicationContext.OutputPath + "kmeans_new.png", kMeansImage.ToUMat());
return(kMeansImage);
}
