solartom: a solar tomography package
🚧🚧🚧 This package is under heavy development and will likely change dramatically. 🚧🚧🚧
🚧🚧🚧 New features are on the way. 🚧🚧🚧
What is solartom ?
Tomography is the task of reconstructing a model based on observations. In this case, we use many 2D solar images to construct a 3D model of the solar environment.
This is a parallelized tomography projector and backprojector. It originates from solar tomography applications but could be used for other applications. It uses the Siddon algorithm as its core. The parallelization is done with Rust's Rayon library.
A similar package called TomograPy was originally authored by Nicolas Barbey. This is the second generation of that package. It is now updated and maintained by Marcus Hughes.
Above you can see an example input with its reconstruction from a solartom derived model cube. It's not perfect but shows the promise of this package.
Status
This package still needs some features and much documentation to improve ease of use for solar physics settings. If you're interested in using it, please contact Marcus Hughes marcus.hughes@swri.org for more information on a timeline.
Installation
Until the code is released on PyPI (coming soon!), you have to clone the repo and then install using pip. I always recommend creating a virtual environment for each project.
git@github.com:jmbhughes/solartom.git
python -m venv venv
source venv/bin/activate
pip install .Basic Use
Right now a simple toy example is available in example.py. More guidance is coming on how to use with STEREO and other data!
Extended Documentation
Coming soon!
Collaborations and questions
Please reach out to Marcus Hughes at marcus.hughes@swri.org. I'd love to have your input and use case in mind when developing this software.