Registry / ai-ml / itk-segmentation

itk-segmentation

library5.4.6pypypi✓ verified 85d ago

ITK is an open-source, cross-platform toolkit for N-dimensional image analysis, segmentation, and registration. The itk-segmentation subpackage provides segmentation algorithms, including level sets, watershed, region growing, and threshold-based methods. Current version 5.4.6 requires Python >=3.8, with ongoing development toward ITK 6.0 (currently in beta). Regular maintenance releases occur approximately quarterly.

ai-ml data

pip install itk-segmentation

github ↗package ↗homepage ↗

INSTALL

IMPORT

SIG · ITK-SEGMENTATION

itk-segmentation

ai-mlpythonv5.4.6

Install

15.9s avg

Import

542ms

Disk

1352MB

Pass rate

5/ 10

Env Coverage5 / 10

glibc

3.9–3.13

musl

3.9–3.13

Install & Compatibility

Where this runs

tested against v5.4.6 · pip install

no network on importno background threads

Install × environment matrix

Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux

musl

py 3.10–3.95 runs

build_error

glibc

py 3.10–3.95 runs

installs and imports cleanly · install 15.9s · import 0.542s · 1331.2MB

1352MB installed

● package 1352MB

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

itk.SegmentationLevelSetImageFilter

✓ import itk

✗ from itkSegmentation import SegmentationLevelSetImageFilter

ITK uses a top-level Python module 'itk' and nested classes; never import subpackages directly.

ImageFileReader

✓ import itk; reader = itk.ImageFileReader[type].New()

✗ from itk import ImageFileReader

ImageFileReader is a template class; you must specify the image type via brackets.

Reads an image, applies a binary threshold, and writes the segmentation.

import itk # Read an image (replace with actual file path or use a sample image) image = itk.imread('input.nii.gz', itk.F) # Apply binary threshold segmentation segmented = itk.binary_threshold_image_filter( image, lower_threshold=100, upper_threshold=500, inside_value=255, outside_value=0 ) # Write result itk.imwrite(segmented, 'segmented.nii.gz')

Debug

Known issues

gotchaITK uses snake_case for simple filter wrappers (e.g., itk.binary_threshold_image_filter) but PascalCase for full filter classes (e.g., itk.BinaryThresholdImageFilter).

fix

For quick scripting use itk.function_name; for advanced control use itk.ClassName.New()

affects: 5.x

breakingITK 6.0 requires C++17 compiler; not all legacy systems have it. Python wheels may not be available for all platforms.

fix

Ensure your build environment supports C++17. Consider using conda or pre-built wheels if available.

affects: >=6.0

deprecatedImageIO factory classes like itk.GDCMImageIO are replaced by itk.GDCMImageIO.New() or automatically detected.

fix

Use itk.imread/imwrite without specifying ImageIO. If needed, use itk.ImageIOFactory.CreateImageIO and set manually.

affects: 5.x

gotchaitk.imread assumes fixed integer types (e.g., itk.UC for unsigned char). For floating point, use itk.F or itk.D.

fix

Explicitly specify pixel type: itk.imread('file', itk.F) reads as float.

affects: 5.x

gotchaTemplate parameters in ITK are specified via brackets: filter = itk.BinaryThresholdImageFilter[InputImageType, OutputImageType].New()

fix

Always provide the image types when using class-based filters; use simple wrappers for convenience.

affects: 5.x

Errors

Common errors & fixes

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

ITK's Python module does not expose all C++ classes directly; some require loading submodules or using different names.

fix

Use functional interface: itk.segmentation_level_set_image_filter(...) or import the specific filter via itk.LevelSetExtension.LevelSetExtensionBase.

RuntimeError: ImportError: libitkcommon.so.6: cannot open shared object file

Missing ITK shared libraries in the system's library path.

fix

Install the 'itk' metapackage or use conda: conda install -c conda-forge itk

TypeError: in method 'New', argument 1 of type 'itk::ImageIOBase::Pointer'

Passing an invalid argument to a filter's New method, often due to wrong template type.

fix

Check that template types and dimensions match. For example: itk.CastImageFilter[InputType, OutputType].New()

ValueError: could not broadcast input array from shape (512,512) into shape (512,512,1)

ITK expects 2D image arrays for 2D images, not 3D with singleton dimension.

fix

Use itk.GetArrayFromImage to get a numpy array, then squeeze: np.squeeze(array) if needed.

Upgrade

Version history

5.4.6latest on PyPI · released Apr 23, 2026

Audit

Dependencies

No dependency data recorded yet.

Agent activity

8 hits · last 30 days

node

Resources

githubhttps://github.com/InsightSoftwareConsortium/ITK ↗

packagehttps://pypi.org/project/itk-segmentation/ ↗

homepagehttps://docs.itk.org/ ↗