https://github.com/dyollb/itkdissolve

ITK external module to dissolve masked regions
https://github.com/dyollb/itkdissolve

c-plus-plus image-processing itk itk-module python segmentation

Last synced: 4 months ago
JSON representation

ITK external module to dissolve masked regions

• Host: GitHub
• URL: https://github.com/dyollb/itkdissolve
• Owner: dyollb
• License: mit
• Created: 2022-08-20T20:11:05.000Z (almost 4 years ago)
• Default Branch: main
• Last Pushed: 2024-03-26T07:10:06.000Z (about 2 years ago)
• Last Synced: 2024-04-26T13:46:27.782Z (about 2 years ago)
• Topics: c-plus-plus, image-processing, itk, itk-module, python, segmentation
• Language: C++
• Homepage:
• Size: 43.9 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# ITKDissolve

[![Build Status](https://github.com/dyollb/ITKDissolve/workflows/Build,%20test,%20package/badge.svg)](https://github.com/dyollb/ITKDissolve/actions)

[![License]( https://img.shields.io/github/license/dyollb/ITKDissolve?color=blue)](https://github.com/dyollb/ITKDissolve/blob/main/LICENSE)

[![PyPI version](https://img.shields.io/pypi/v/itk-dissolve.svg)](https://badge.fury.io/py/itk-dissolve)





## Overview

This is a module for the Insight Toolkit ([ITK](https://github.com/InsightSoftwareConsortium/ITK)) that provides functionality to discard pixels within a masked region. Unlike masking, the pixels in the masked region are replaced (dissolved) by their nearest pixels outside the mask. This can be useful to clean-up label fields, e.g. to remove small islands/holes or remove entire labels, and replace them by the adjacent labels.

The module includes a filter called DissolveMaskImageFilter.

```python

    import itk

    labels = itk.imread('path/to/labels.mha').astype(itk.US)

    mask = itk.imread('path/to/mask.mha').astype(itk.UC)

    ImageType = type(labels)

    MaskType = type(mask)

    dissolve = itk.DissolveMaskImageFilter[ImageType, MaskType].New()

    dissolve.SetInput(labels)

    dissolve.SetMaskImage(mask)

    dissolve.Update()

    modified_labels = dissolve.GetOutput()

    itk.imwrite(modified_labels, 'modified_labels2.mha')

```

Or using the pythonic API:

```python

    import itk

    labels = itk.imread('path/to/labels.mha').astype(itk.US)

    mask = itk.imread('path/to/mask.mha').astype(itk.US)

    modified_labels = itk.dissolve_mask_image_filter(labels, mask_image=mask)

    itk.imwrite(modified_labels, 'path/to/modified_labels.mha')

```

## Installation

To install the binary Python packages:

```shell

  python -m pip install itk-dissolve

```