The CTSimU Software Toolbox is a set of Python modules to run several image processing and analysis routines that are common in computed tomography (CT).

Features

The toolbox provides the following main submodules, each designed for a different task.

Quick Access

The ctsimu.toolbox sub-module provides a simplified interface to the most commonly used functionality. It reads CTSimU scenario description files and metadata files to carry out the following tasks:

• Flat-field correction,
• Create reconstruction configuration files,
• Standardize or update scenario description files to the latest file format version,
• Run recursive post-processing on whole directory trees (batch processing).

Image conversion & processing

The ctsimu.processing sub-module provides image processing pipelines that can be used to convert stacks of projection images into different formats. The following properties can be converted:

• Image format conversion (RAW, TIFF),
• Data type conversion (int, float),
• Representation conversion (multi-file slice stacks, single-file RAW volume chunks).

Additional processing steps can be added to a pipeline: this will alter the images before they are saved. Currently, there are processing steps for

• Flat-field and dark-field correction with provided gain/offset files,
• Pixel binning,
• Median filter,
• Transformations: rotations, flipping.

Image Quality Assessment

The ctsimu.image_analysis sub-module currently provides functions for:

• Line profiles,
• ctsimu.image_analysis.mtf: Calculation of the modulation transfer function (MTF) from an edge image,
• ctsimu.image_analysis.isrb: Determination of the interpolated basic spatial resolution (iSRb) from a duplex wire image.

Geometry Tools

The ctsimu.geometry sub-module provides tools for

• Geometric representation and manipulation of a CT scan,
• Calculation of projection matrices,
• Creation of reconstruction configurations for openCT (VGSTUDIO) and CERA, even though it is easier to use the ctsimu.scenario sub-module for this.

Scenario Handling

The ctsimu.scenario sub-module is used to handle complete CTSimU scenarios:

• Read, write, explore and manipulate CTSimU scenario description files,
• Calculate coordinate systems for each frame,
• Generate reconstruction configuration files.

2D Test Evaluations

The ctsimu.evaluation sub-module provides scripts and documentation for CTSimU 2D test evaluations.

Getting started

Requirements

A Python 3 environment (Python 3.9 or higher) is required. The Python distribution that was used to develop and test the toolbox is Anaconda 3 (available for Windows, macOS and Linux), and therefore a recommendation if you don't know where to start. The following Python packages are required as well. They usually come with a Python environment, or can be installed there easily:

• NumPy,
• SciPy subpackages ndimage, optimize, stats, signal and
• Matplotlib (only required if you want plots of evaluation results).

Installation

If you want to install the package in your Python environment, you can use pip. For example, you can run the following command in your Anaconda Prompt to make the toolbox available:

pip install ctsimu

To use the package without installation, you need to download the package manually. You have the following three options:

• Download the package from PyPi. You will get a zipped file called ctsimu-X.X.X.tar.gz (where X.X.X is the current version number). Open or unpack this file to get to its contents.
• Download the repository from GitHub: press the clone button to download a ZIP file. Unpack it or open it to see its contents.

• You can also clone the repository from GitHub via the terminal:

  git clone https://github.com/BAMresearch/ctsimu-toolbox.git

From any of these three options, you will get the complete package source code. It contains a folder called ctsimu. If you want to use the toolbox from your Python scripts without installing it, you need to copy the folder ctsimu to the location of your Python scripts to make the package available.

About

The CTSimU toolbox was developed for the CTSimU project for its needs in reproducible image processing and serves as the project's reference software implementation. It provides useful tools for many CT-related tasks, beyond the aims of the CTSimU project.

The software is released under the Apache 2.0 license, its source code is available on GitHub.

Acknowledgements

This work was funded through the project CTSimU (Radiographic Computed Tomography Simulation for Measurement Uncertainty Evaluation, WIPANO project 03TNH026A).

WIPANO projects are financed by the German Federal Ministry for Economic Affairs and Climate Action and managed by Project Management Jülich.

Contributors

The following people contributed to code and documentation of the toolbox.

• David Plotzki
• Bendix Hartlaub
• Florian Wohlgemuth
• Tamara Reuter
• Fabrício Borges de Oliveira
• Carsten Bellon

# -*- coding: UTF-8 -*-
"""The CTSimU Software Toolbox is a set of Python modules to run several image processing and analysis routines that are common in computed tomography (CT).

.. include:: ./documentation.md
"""

__pdoc__ = {}

Sub-modules

ctsimu.evaluation

Evaluations for CTSimU 2D tests …

ctsimu.geometry

Coordinate systems, transformations and projection matrix functionality …

ctsimu.helpers

ctsimu.image

Single image handling and processing …

ctsimu.image_analysis

ctsimu.primitives

Structures for linear algebra and geometry: vectors, matrices, lines and polygons.

ctsimu.processing

Image batch processing tools …

ctsimu.scenario

Read, write, handle and manipulate a CTSimU scenario …

ctsimu.test

General class for a projection-based test.

ctsimu.tiffy

TIFF image reader and writer.

ctsimu.toolbox

Quick metadata-based post-processing with simple commands …