Floodfill Python ITK

Algorithms
1

There’s a function in python to floodfill an image that has been previously segmented?

2

What do you mean by flood fill? Does ConnectedComponent filter do what you want?

3

I’m trying to color different regions defined by edges of an image (different regions inside that image).

4

Sounds like a job for LabelToRGBImageFilter. You might need to segment your image first, which might be easy, hard, or anything in between.

5

Thank you so much for your support

6

I tried to apply a custom color map in my code but i have an error that i don’t understand. Do you have any ideas? I want to apply this Custom mask to my code, because i want to watershed it, after the previous segmentation
This is a little explication about what i have to do
Cattura

import sys
import itk
inputFileName = “C:\Users\fraca\Desktop\IM21”
outputFileName = “C:\Users\fraca\Desktop\zerocrossing.mha”
Dimension = 2
FloatPixelType = itk.ctype(‘float’)
FloatImageType = itk.Image[FloatPixelType, Dimension]
reader = itk.ImageFileReader[FloatImageType].New()
reader.SetFileName(inputFileName)
zero=itk.ZeroCrossingBasedEdgeDetectionImageFilter.New(input=reader.GetOutput())
watershed = itk.WatershedImageFilter.New(Input=zero.GetOutput())
LabeledImageType = type(watershed.GetOutput())
PixelType = itk.ctype(‘unsigned char’)
RGBPixelType = itk.RGBPixel[PixelType]
RGBImageType = itk.Image[RGBPixelType, Dimension]
ColormapType = itk.CustomColormapFunction[PixelType, RGBPixelType]
colormap = ColormapType.New()
random=itk.MersenneTwisterRandomVariateGenerator.New()
random.SetSeed(0)
redChannel=
greenChannel=
blueChannel=
for i in range(255): {
redChannel.append(random.GetUniformVariate(0.0, 1.0)) greenChannel.append(random.GetUniformVariate(0.0, 1.0)) blueChannel.append(random.GetUniformVariate(0.0, 1.0))
}
colormap.SetRedChannel(redChannel)
colormap.SetGreenChannel(greenChannel)
colormap.SetBlueChannel(blueChannel)
ScalarToRGBColormapFilterType=itk.ScalarToRGBColormapImageFilter[LabeledImageType,RGBImageType]
colormapImageFilter=ScalarToRGBColormapFilterType.New()
colormapImageFilter.SetColormap(colormap)#colormap
colormapImageFilter.SetInput(watershed.GetOutput())
colormapImageFilter.Update()
WriterType = itk.ImageFileWriter[RGBImageType]
writer = WriterType.New()
writer.SetFileName(outputFileName)
writer.SetInput(colormapImageFilter.GetOutput())
writer.Update()

—> 55 colormapImageFilter.SetColormap(colormap)
56 colormapImageFilter.SetInput(watershed.GetOutput())
57 #colormapImageFilter.Update()
TypeError: in method ‘itkScalarToRGBColormapImageFilterIULL2IRGBUC2_SetColormap’, argument 2 of type ‘itkScalarToRGBColormapImageFilterIULL2IRGBUC2::ColormapEnumType’

7

This example, than uses the MorphologicalWatershedImageFilter and ScalarToRGBColormapImageFilter may be helpful.

Note that the MorphologicalWatershedImageFilter behaves much better than the WatershedImageFilter, and it is preferred.

2 Likes
8

Can you help me to convert these rows of code in python? I tried something but i always obtain some errors

binaryHeavyGaussianFilterX->SetOrder(RecursiveGaussianFilterType::ZeroOrder);
binaryHeavyGaussianFilterY->SetOrder(RecursiveGaussianFilterType::ZeroOrder);

and

if (vcl_fabs(denominator) < epsilon)
{
denominator = denominator > itk::NumericTraits< PixelType >::Zero ? 1.0 : -1.0;
}

9

Here is the second part:

if abs(denominator) < epsilon
  if denominator > 0
    denominator = 1.0
  else
    denominator = -1.0
2 Likes
10

This will be available in Python with ITK 5.1 RC 1 as:

binaryHeavyGaussianFilterX.SetOrder(itk.GaussianOrderEnum_ZeroOrder)
2 Likes