Stacking Z dimension images using SimpleITK

Beginner Questions
,
1

I’m trying to stack multiple Z dimension images where the images are taken in XYZ axis. (Images are in .tiff format) I want those 2D images to create a 3D Volume.

I was tinkering with SimpleITK on Python, I got some results but it wasn’t good. It looked like merged images into one. Rather creating a 3D cell-volume. (I was using Pyvista library for visualisation)

I think it’s achievable in SimpleITK but unsure how. Can someone point me to the right direction? Thank you!

(I’m working on mice brain scans)

2

This is a cross-post from image.sc forum.

3

Hello @sleepingcat4,

The relevant classes are ImageSeriesReader and JoinSeriesImageFilter.

Sample code:

import SimpleITK as sitk

num_images_in_z = 5
# all z slices are expected to have the same number of pixels, origin, spacing and
# direction cosine matrix. The code below creates num_images_in_z accordingly and
# each has a unique constant value
all_z_slices = [sitk.Image([512,512], sitk.sitkUInt16)+i for i in range(num_images_in_z)]
volume = sitk.JoinSeries(all_z_slices)
sitk.Show(volume)
4

Yes, I thought SimpleITK didn’t have dedicated forum. When I saw ITK had its dedicated forum, I posted my question here as well : )

5

Thank you for the help! I wanted to add another detail: my images look like this

image
image1647×657 154 KB

although obviously not all the images for the stack to be complete are present here. But, Can you guide me what method I should use to make this images transform from 2D to 3D Volume using SimpleITK.

Example code will be really helpful for me to get started + if you can point me to the right documentation as well. @zivy

What I already tried?
I tried this below code

import os
import SimpleITK as sitk
import cv2
import numpy as np

folder_path = "sample"
file_names = os.listdir(folder_path)
file_names.sort()
image_list = []

for file_name in file_names:
    image_path = os.path.join(folder_path, file_name)
    
    # Use OpenCV to read the TIFF image
    image_cv2 = cv2.imread(image_path, cv2.IMREAD_UNCHANGED)
    
    image = sitk.GetImageFromArray(image_cv2)
    image_list.append(image)

volume = sitk.JoinSeries(image_list)
output_filename = "stacked_volume.nii.gz"
output_path = os.path.join(os.getcwd(), output_filename)
sitk.WriteImage(volume, output_path)

later used Pyvista library to visualise the results. It didn’t give me a 3D Volume rather it was a static image merged in 3D axis. I would appreciate all the help you can give me. Thank you!

6

Hello @sleepingcat4,

The code is overly complicated. ITK/SimpleITK is able to read TIFF images. If you provide the list of files sorted according to z they will be readily read as a 3D image (volume). Not sure what you mean by “static image merged in 3D axis”. If these images are supposed to be a single 2D image with multiple channels that is one line of code too, just a different one. See code below:

import os
import SimpleITK as sitk

folder_path = "sample"
file_names = os.listdir(folder_path)
file_names.sort()

# if files are supposed to be a 3D image
image = sitk.ReadImage(file_names.sort())

# or if files are supposed to be a 2D multi channel image use the line below
#image = sitk.Compose([sitk.ReadImage(file_name) for file_name in file_names])

# check if the spatial information was read correctly
print(f"image size: {image.GetSize()}")
print(f"image spacing: {image.GetSpacing()}")
print(f"image origin: {image.GetOrigin()}")
2 Likes
7

Thanks for the help! I modified your code a bit to prevent NonType exception. and these are the results I received:

import os
import SimpleITK as sitk

folder_path = "sample"
file_names = os.listdir(folder_path)
file_names.sort()  # Sort the list of file names

# If files are supposed to be a 3D image
image = sitk.ReadImage([os.path.join(folder_path, file_name) for file_name in file_names])

# Check if the spatial information was read correctly
print(f"Image size: {image.GetSize()}")
print(f"Image spacing: {image.GetSpacing()}")
print(f"Image origin: {image.GetOrigin()}")

-------------------
results

Image size: (2048, 2048, 8)
Image spacing: (1.0, 1.0, 1.0)
Image origin: (0.0, 0.0, 0.0)

Can you guide me what route I should take to stack those XYZ dimension 2D images into 3D Volume?

@zivy

8

Hello @sleepingcat4,

The images were stacked into a volume, it’s just a very thin one, 8 slices in z direction vs. the 2048 for x and y. One thing that is suspicious is the image spacing. It is rarely, if ever, [1,1,1].

Is the pixel size information available in the tiff images? If yes, then there is a problem in the reading. The x,y,z spacing of [1,1,1] is a default used when the pixel spacing information is missing or not detected. If you know the original spacings then you can set them image.SetSpacing([x_spacing, y_spacing, z_spacing]) and then the volume will have the correct size.

If you don’t know whether the files contain the spatial information you could try another package that can read tiff files (a different implementation) see what that reports. Specifically pyvips:

import pyvips
file_name = 
image = pyvips.Image.new_from_file(file_name)
print(f"spacing: {image.xres}, {image.yres}")
9

I have used the code you provided earlier to both stack the images together for a 3D Volume construction and also to see the spacing. After stacking the images together using the following code, I received this result:

import os
import SimpleITK as sitk

folder_path = "sample"
file_names = os.listdir(folder_path)
file_names.sort()  # Sort the list of file names

# If files are supposed to be a 3D image
# image = sitk.ReadImage([os.path.join(folder_path, file_name) for file_name in file_names])

# Check if the spatial information was read correctly
# print(f"Image size: {image.GetSize()}")
# print(f"Image spacing: {image.GetSpacing()}")
# print(f"Image origin: {image.GetOrigin()}")

image = sitk.Compose([sitk.ReadImage(os.path.join(folder_path, file_name)) for file_name in file_names])

# saving
output_file_path = "result.nii.gz"
sitk.WriteImage(image, output_file_path)

print(f"Result saved as {output_file_path}")

image
image917×851 23.9 KB

Yes, I realise the layers are extremely thin, still I am unsure why it looks like a 2D merged image than a crude 3D construction? If you could explain on that, will be helpful.

I used this below code and Pyvista to visualise the results

import pyvista as pv

# Load the file as a PyVista grid
grid = pv.read("result.nii.gz")

# Create a plotter and add the grid
plotter = pv.Plotter()
plotter.add_mesh(grid, opacity=0.5)

plotter.show()

Also after running the pyvips library, I received this result

import pyvips
file_name = "Image_T0000_C0000_X0000_Y0000_Z1000.tiff"
image = pyvips.Image.new_from_file(file_name)
print(f"spacing: {image.xres}, {image.yres}")

-----------------------------
result
spacing: 1.0, 1.0

Btw the title configuration register file have these following values

# Define the number of dimensions we are working on
dim = 2

# Define the image coordinates
Image_T0000_C0000_X0000_Y0000_Z1000.tiff; ; (0.0, 0.0)
Image_T0000_C0000_X0001_Y0000_Z1000.tiff; ; (1544.9514162524083, 17.072951450086833)
Image_T0000_C0000_X0002_Y0000_Z1000.tiff; ; (3096.9032049704, 37.14534362046079)
Image_T0000_C0000_X0003_Y0000_Z1000.tiff; ; (4643.90301598235, 54.14568694049926)

And the tile configuration file contains these values

# Define the number of dimensions we are working on
dim = 2

# Define the image coordinates
Image_T0000_C0000_X0000_Y0000_Z1000.tiff; ; (0.0, 0.0)
Image_T0000_C0000_X0001_Y0000_Z1000.tiff; ; (1638.0, 0.0)
Image_T0000_C0000_X0002_Y0000_Z1000.tiff; ; (3276.0, 0.0)

If you can help me out, that’ll be helpful thanks for the previous help as well :slight_smile:

@zivy

10

If you have a 2D TileConfiguration.txt, you might not have a 3D stack - you might have a 2D montage. Take a look at this example, and/or this one.

2 Likes
11

I’ll take a look and run some code to see if it reaches my desired outcome. And let you guys know

12

I’m getting error, I think SimpleITK is changed a lot since the code was written. Can you guys please help me out?

I received this errors while working with the code shown in this example

errors, I received

AttributeError                            Traceback (most recent call last)
Cell In [6], line 5
      1 dimension = 2
      3 tile_config_file = "TawsifExampleDataApr2024\RawTiles\FOS_TMP_1000\TileConfiguration.txt"
----> 5 stage_tiles = itk.TileConfiguration[dimension]()
      6 stage_tiles.Parse(tile_config_file)

AttributeError: module 'SimpleITK' has no attribute 'TileConfiguration'

also received another error

AttributeError                            Traceback (most recent call last)
Cell In [7], line 7
      4 import SimpleITK as itk
      5 import matplotlib.pyplot as plt
----> 7 itk.auto_progress(1)
      8 itk.TileConfiguration.GetTypes()
      9 itk.auto_progress(0)

AttributeError: module 'SimpleITK' has no attribute 'auto_progress'

I’m new to this library, if guys could help me out, that’ll be really helpful. @dzenanz @zivy

13

Hello @sleepingcat4,

The ITKMontage module is part of ITK but not part of SimpleITK. You will need to switch to using ITK (pip install itk or conda install -c conda-forge itk depending on which Python you are using).

1 Like
14

thanks for the help. I have a new problem:

AttributeError                            Traceback (most recent call last)
Cell In [1], line 9
      7 # Pre-load ITKMontage components
      8 itk.auto_progress(1)
----> 9 itk.TileConfiguration.GetTypes()
     10 itk.auto_progress(0)

File ~\AppData\Roaming\Python\Python38\site-packages\itk\support\lazy.py:131, in LazyITKModule.__getattribute__(self, attr)
    130 def __getattribute__(self, attr):
--> 131     value = types.ModuleType.__getattribute__(self, attr)
    132     if value is not_loaded:
    133         with type(self).get_lock():  # All but one thread will block here.

AttributeError: module 'itk' has no attribute 'TileConfiguration'

I’m unsure whether itk removed this TileConfiguration module or not. Can you help me out? @zivy

15

You need pip install itk-montage.

1 Like
16

thanks for kind help. I am getting an error where it says, the filter module is receiving two different input type whilst the previous code blocks from the example ran just fine. Can you help me out with this?

Code producing the error:

print("Producing the mosaic")
resampleF = itk.TileMergeImageFilter[type(color_images[0]), itk.RGBPixel[itk.F]].New()
resampleF.SetMontageSize(stage_tiles.GetAxisSizes())
for t in range(stage_tiles.LinearSize()):
    resampleF.SetInputTile(t, color_images[t])
    index = stage_tiles.LinearIndexToNDIndex(t)
    resampleF.SetTileTransform(index, montage.GetOutputTransform(index))
resampleF.Update()
itk.imwrite(resampleF.GetOutput(), str(out_file))
print("Resampling complete")

error:

{
	"name": "TemplateTypeError",
	"message": "itk.TileMergeImageFilter is not wrapped for input type `itk.Image[itk.UC,2], itk.RGBPixel[itk.F]`.

To limit the size of the package, only a limited number of
types are available in ITK Python. To print the supported
types, run the following command in your python environment:

    itk.TileMergeImageFilter.GetTypes()

Possible solutions:
* If you are an application user:
** Convert your input image into a supported format (see below).
** Contact developer to report the issue.
* If you are an application developer, force input images to be
loaded in a supported pixel type.

    e.g.: instance = itk.TileMergeImageFilter[itk.Image[itk.SS,2], itk.D].New(my_input)

* (Advanced) If you are an application developer, build ITK Python yourself and
turned to `ON` the corresponding CMake option to wrap the pixel type or image
dimension you need. When configuring ITK with CMake, you can set
`ITK_WRAP_${type}` (replace ${type} with appropriate pixel type such as
`double`). If you need to support images with 4 or 5 dimensions, you can add
these dimensions to the list of dimensions in the CMake variable
`ITK_WRAP_IMAGE_DIMS`.

Supported input types:

itk.Image[itk.SS,2]
itk.Image[itk.UC,2]
itk.Image[itk.US,2]
itk.Image[itk.F,2]
itk.Image[itk.D,2]
itk.Image[itk.RGBPixel[itk.UC],2]
itk.Image[itk.RGBAPixel[itk.UC],2]
itk.Image[itk.SS,3]
itk.Image[itk.UC,3]
itk.Image[itk.US,3]
itk.Image[itk.F,3]
itk.Image[itk.D,3]
itk.Image[itk.RGBPixel[itk.UC],3]
itk.Image[itk.RGBAPixel[itk.UC],3]
itk.Image[itk.SS,4]
itk.Image[itk.UC,4]
itk.Image[itk.US,4]
itk.Image[itk.F,4]
itk.Image[itk.D,4]
itk.Image[itk.RGBPixel[itk.UC],4]
itk.Image[itk.RGBAPixel[itk.UC],4]
",
	"stack": "---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
File ~\\AppData\\Roaming\\Python\\Python38\\site-packages\\itk\\support\\template_class.py:525, in itkTemplate.__getitem__(self, parameters)
    524 try:
--> 525     this_item = self.__template__[key]
    526 except KeyError:

KeyError: (<class 'itk.itkImagePython.itkImageUC2'>, <class 'itk.itkRGBPixelPython.itkRGBPixelF'>)

During handling of the above exception, another exception occurred:

TemplateTypeError                         Traceback (most recent call last)
Cell In [11], line 2
      1 print(\"Producing the mosaic\")
----> 2 resampleF = itk.TileMergeImageFilter[type(color_images[0]), itk.RGBPixel[itk.F]].New()
      3 resampleF.SetMontageSize(stage_tiles.GetAxisSizes())
      4 for t in range(stage_tiles.LinearSize()):

File ~\\AppData\\Roaming\\Python\\Python38\\site-packages\\itk\\support\\template_class.py:529, in itkTemplate.__getitem__(self, parameters)
    526 except KeyError:
    527     import itk
--> 529     raise itk.TemplateTypeError(self, key)
    530 return this_item

TemplateTypeError: itk.TileMergeImageFilter is not wrapped for input type `itk.Image[itk.UC,2], itk.RGBPixel[itk.F]`.

To limit the size of the package, only a limited number of
types are available in ITK Python. To print the supported
types, run the following command in your python environment:

    itk.TileMergeImageFilter.GetTypes()

Possible solutions:
* If you are an application user:
** Convert your input image into a supported format (see below).
** Contact developer to report the issue.
* If you are an application developer, force input images to be
loaded in a supported pixel type.

    e.g.: instance = itk.TileMergeImageFilter[itk.Image[itk.SS,2], itk.D].New(my_input)

* (Advanced) If you are an application developer, build ITK Python yourself and
turned to `ON` the corresponding CMake option to wrap the pixel type or image
dimension you need. When configuring ITK with CMake, you can set
`ITK_WRAP_${type}` (replace ${type} with appropriate pixel type such as
`double`). If you need to support images with 4 or 5 dimensions, you can add
these dimensions to the list of dimensions in the CMake variable
`ITK_WRAP_IMAGE_DIMS`.

Supported input types:

itk.Image[itk.SS,2]
itk.Image[itk.UC,2]
itk.Image[itk.US,2]
itk.Image[itk.F,2]
itk.Image[itk.D,2]
itk.Image[itk.RGBPixel[itk.UC],2]
itk.Image[itk.RGBAPixel[itk.UC],2]
itk.Image[itk.SS,3]
itk.Image[itk.UC,3]
itk.Image[itk.US,3]
itk.Image[itk.F,3]
itk.Image[itk.D,3]
itk.Image[itk.RGBPixel[itk.UC],3]
itk.Image[itk.RGBAPixel[itk.UC],3]
itk.Image[itk.SS,4]
itk.Image[itk.UC,4]
itk.Image[itk.US,4]
itk.Image[itk.F,4]
itk.Image[itk.D,4]
itk.Image[itk.RGBPixel[itk.UC],4]
itk.Image[itk.RGBAPixel[itk.UC],4]
"
}

reference to the example: ITKMontage/examples/Simple Montage.ipynb at master · InsightSoftwareConsortium/ITKMontage · GitHub

Imagetype for my images are

<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>
<class 'itk.itkImagePython.itkImageUC2'>

Pixel type

Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>
Pixel Type: <itkCType unsigned char>

@dzenanz @zivy

17

Your images are grayscale, so filter instantiation should be like:

resampleF = itk.TileMergeImageFilter[type(color_images[0]), itk.F].New()

or like

resampleF = itk.TileMergeImageFilter[type(color_images[0])].New()
18

Still, I am receiving an error for both of the fixes you suggested.

{
	"name": "TemplateTypeError",
	"message": "itk.TileMergeImageFilter is not wrapped for input type `itk.Image[itk.UC,2], itk.F`.

To limit the size of the package, only a limited number of
types are available in ITK Python. To print the supported
types, run the following command in your python environment:

    itk.TileMergeImageFilter.GetTypes()

Possible solutions:
* If you are an application user:
** Convert your input image into a supported format (see below).
** Contact developer to report the issue.
* If you are an application developer, force input images to be
loaded in a supported pixel type.

    e.g.: instance = itk.TileMergeImageFilter[itk.Image[itk.SS,2], itk.D].New(my_input)

* (Advanced) If you are an application developer, build ITK Python yourself and
turned to `ON` the corresponding CMake option to wrap the pixel type or image
dimension you need. When configuring ITK with CMake, you can set
`ITK_WRAP_${type}` (replace ${type} with appropriate pixel type such as
`double`). If you need to support images with 4 or 5 dimensions, you can add
these dimensions to the list of dimensions in the CMake variable
`ITK_WRAP_IMAGE_DIMS`.

Supported input types:

itk.Image[itk.SS,2]
itk.Image[itk.UC,2]
itk.Image[itk.US,2]
itk.Image[itk.F,2]
itk.Image[itk.D,2]
itk.Image[itk.RGBPixel[itk.UC],2]
itk.Image[itk.RGBAPixel[itk.UC],2]
itk.Image[itk.SS,3]
itk.Image[itk.UC,3]
itk.Image[itk.US,3]
itk.Image[itk.F,3]
itk.Image[itk.D,3]
itk.Image[itk.RGBPixel[itk.UC],3]
itk.Image[itk.RGBAPixel[itk.UC],3]
itk.Image[itk.SS,4]
itk.Image[itk.UC,4]
itk.Image[itk.US,4]
itk.Image[itk.F,4]
itk.Image[itk.D,4]
itk.Image[itk.RGBPixel[itk.UC],4]
itk.Image[itk.RGBAPixel[itk.UC],4]
",
	"stack": "---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
File ~\\AppData\\Roaming\\Python\\Python38\\site-packages\\itk\\support\\template_class.py:525, in itkTemplate.__getitem__(self, parameters)
    524 try:
--> 525     this_item = self.__template__[key]
    526 except KeyError:

KeyError: (<class 'itk.itkImagePython.itkImageUC2'>, <itkCType float>)

During handling of the above exception, another exception occurred:

TemplateTypeError                         Traceback (most recent call last)
Cell In [10], line 2
      1 print(\"Producing the mosaic\")
----> 2 resampleF = itk.TileMergeImageFilter[type(color_images[0]), itk.F].New()#  line creating the problem
      3 resampleF.SetMontageSize(stage_tiles.GetAxisSizes())
      4 for t in range(stage_tiles.LinearSize()):

File ~\\AppData\\Roaming\\Python\\Python38\\site-packages\\itk\\support\\template_class.py:529, in itkTemplate.__getitem__(self, parameters)
    526 except KeyError:
    527     import itk
--> 529     raise itk.TemplateTypeError(self, key)
    530 return this_item

TemplateTypeError: itk.TileMergeImageFilter is not wrapped for input type `itk.Image[itk.UC,2], itk.F`.

To limit the size of the package, only a limited number of
types are available in ITK Python. To print the supported
types, run the following command in your python environment:

    itk.TileMergeImageFilter.GetTypes()

Possible solutions:
* If you are an application user:
** Convert your input image into a supported format (see below).
** Contact developer to report the issue.
* If you are an application developer, force input images to be
loaded in a supported pixel type.

    e.g.: instance = itk.TileMergeImageFilter[itk.Image[itk.SS,2], itk.D].New(my_input)

* (Advanced) If you are an application developer, build ITK Python yourself and
turned to `ON` the corresponding CMake option to wrap the pixel type or image
dimension you need. When configuring ITK with CMake, you can set
`ITK_WRAP_${type}` (replace ${type} with appropriate pixel type such as
`double`). If you need to support images with 4 or 5 dimensions, you can add
these dimensions to the list of dimensions in the CMake variable
`ITK_WRAP_IMAGE_DIMS`.

Supported input types:

itk.Image[itk.SS,2]
itk.Image[itk.UC,2]
itk.Image[itk.US,2]
itk.Image[itk.F,2]
itk.Image[itk.D,2]
itk.Image[itk.RGBPixel[itk.UC],2]
itk.Image[itk.RGBAPixel[itk.UC],2]
itk.Image[itk.SS,3]
itk.Image[itk.UC,3]
itk.Image[itk.US,3]
itk.Image[itk.F,3]
itk.Image[itk.D,3]
itk.Image[itk.RGBPixel[itk.UC],3]
itk.Image[itk.RGBAPixel[itk.UC],3]
itk.Image[itk.SS,4]
itk.Image[itk.UC,4]
itk.Image[itk.US,4]
itk.Image[itk.F,4]
itk.Image[itk.D,4]
itk.Image[itk.RGBPixel[itk.UC],4]
itk.Image[itk.RGBAPixel[itk.UC],4]
"
}

for the other one

{
	"name": "TemplateTypeError",
	"message": "itk.TileMergeImageFilter is not wrapped for input type `itk.Image[itk.UC,2]`.

To limit the size of the package, only a limited number of
types are available in ITK Python. To print the supported
types, run the following command in your python environment:

    itk.TileMergeImageFilter.GetTypes()

Possible solutions:
* If you are an application user:
** Convert your input image into a supported format (see below).
** Contact developer to report the issue.
* If you are an application developer, force input images to be
loaded in a supported pixel type.

    e.g.: instance = itk.TileMergeImageFilter[itk.Image[itk.SS,2], itk.D].New(my_input)

* (Advanced) If you are an application developer, build ITK Python yourself and
turned to `ON` the corresponding CMake option to wrap the pixel type or image
dimension you need. When configuring ITK with CMake, you can set
`ITK_WRAP_${type}` (replace ${type} with appropriate pixel type such as
`double`). If you need to support images with 4 or 5 dimensions, you can add
these dimensions to the list of dimensions in the CMake variable
`ITK_WRAP_IMAGE_DIMS`.

Supported input types:

itk.Image[itk.SS,2]
itk.Image[itk.UC,2]
itk.Image[itk.US,2]
itk.Image[itk.F,2]
itk.Image[itk.D,2]
itk.Image[itk.RGBPixel[itk.UC],2]
itk.Image[itk.RGBAPixel[itk.UC],2]
itk.Image[itk.SS,3]
itk.Image[itk.UC,3]
itk.Image[itk.US,3]
itk.Image[itk.F,3]
itk.Image[itk.D,3]
itk.Image[itk.RGBPixel[itk.UC],3]
itk.Image[itk.RGBAPixel[itk.UC],3]
itk.Image[itk.SS,4]
itk.Image[itk.UC,4]
itk.Image[itk.US,4]
itk.Image[itk.F,4]
itk.Image[itk.D,4]
itk.Image[itk.RGBPixel[itk.UC],4]
itk.Image[itk.RGBAPixel[itk.UC],4]
",
	"stack": "---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
File ~\\AppData\\Roaming\\Python\\Python38\\site-packages\\itk\\support\\template_class.py:525, in itkTemplate.__getitem__(self, parameters)
    524 try:
--> 525     this_item = self.__template__[key]
    526 except KeyError:

KeyError: (<class 'itk.itkImagePython.itkImageUC2'>,)

During handling of the above exception, another exception occurred:

TemplateTypeError                         Traceback (most recent call last)
Cell In [11], line 2
      1 print(\"Producing the mosaic\")
----> 2 resampleF = itk.TileMergeImageFilter[type(color_images[0])].New() #  line creating the problem
      3 resampleF.SetMontageSize(stage_tiles.GetAxisSizes())
      4 for t in range(stage_tiles.LinearSize()):

File ~\\AppData\\Roaming\\Python\\Python38\\site-packages\\itk\\support\\template_class.py:529, in itkTemplate.__getitem__(self, parameters)
    526 except KeyError:
    527     import itk
--> 529     raise itk.TemplateTypeError(self, key)
    530 return this_item

TemplateTypeError: itk.TileMergeImageFilter is not wrapped for input type `itk.Image[itk.UC,2]`.

To limit the size of the package, only a limited number of
types are available in ITK Python. To print the supported
types, run the following command in your python environment:

    itk.TileMergeImageFilter.GetTypes()

Possible solutions:
* If you are an application user:
** Convert your input image into a supported format (see below).
** Contact developer to report the issue.
* If you are an application developer, force input images to be
loaded in a supported pixel type.

    e.g.: instance = itk.TileMergeImageFilter[itk.Image[itk.SS,2], itk.D].New(my_input)

* (Advanced) If you are an application developer, build ITK Python yourself and
turned to `ON` the corresponding CMake option to wrap the pixel type or image
dimension you need. When configuring ITK with CMake, you can set
`ITK_WRAP_${type}` (replace ${type} with appropriate pixel type such as
`double`). If you need to support images with 4 or 5 dimensions, you can add
these dimensions to the list of dimensions in the CMake variable
`ITK_WRAP_IMAGE_DIMS`.

Supported input types:

itk.Image[itk.SS,2]
itk.Image[itk.UC,2]
itk.Image[itk.US,2]
itk.Image[itk.F,2]
itk.Image[itk.D,2]
itk.Image[itk.RGBPixel[itk.UC],2]
itk.Image[itk.RGBAPixel[itk.UC],2]
itk.Image[itk.SS,3]
itk.Image[itk.UC,3]
itk.Image[itk.US,3]
itk.Image[itk.F,3]
itk.Image[itk.D,3]
itk.Image[itk.RGBPixel[itk.UC],3]
itk.Image[itk.RGBAPixel[itk.UC],3]
itk.Image[itk.SS,4]
itk.Image[itk.UC,4]
itk.Image[itk.US,4]
itk.Image[itk.F,4]
itk.Image[itk.D,4]
itk.Image[itk.RGBPixel[itk.UC],4]
itk.Image[itk.RGBAPixel[itk.UC],4]
"
}

Can you help me with this? @dzenanz

19

Try resampleF = itk.TileMergeImageFilter[type(color_images[0]), itk.D].New(). I don’t know why the second suggestion (without explicit second parameter) is not working. itk.Image[itk.UC,2] is on the list of supported input types.

1 Like
20

thanks this works but I had an question and problem. Why the resulted image has waves? is there a way to prevent this? @dzenanz

image

1 Like