10.5281/zenodo.5221443
https://zenodo.org/records/5221443
oai:zenodo.org:5221443
Kristian Bredies
Kristian Bredies
0000-0001-7140-043X
University of Graz
Richard Huber
Richard Huber
0000-0003-1743-6786
University of Graz
Gratopy 0.1 release candidate 1
Zenodo
2021
Radon transform
fanbeam transform
pixel-driven projection methods
computed tomography
image reconstruction
2021-08-19
eng
https://github.com/kbredies/gratopy/tree/v0.1.0-beta.0
10.5281/zenodo.5221442
v0.1.0-beta.0
The Graz accelerated tomographic projection for Python (Gratopy) is a software tool for Python3 developed to allow for efficient, high quality execution of projection methods such as Radon and fanbeam transform. The operations contained in the toolbox are based on pixel-driven projection methods, which were shown to possess suitable approximation properties. The code is based in a powerful OpenCL/GPU implementation, resulting in high execution speed, while allowing for seamless integration into PyOpenCL. Hence this can efficiently be paired with other PyOpenCL code, in particular OpenCL based optimization algorithms.
Highlights
Easy-to-use tomographic projection toolbox.
High-quality 2D projection operators.
Fast projection due to custom OpenCL/GPU implementation.
Seamless integration into PyOpenCL.
Basic iterative reconstruction schemes included (Landweber, CG, total variation).
Comprehensive documentation, tests and example code.
Note that this is a pre-release.
FWF Austrian Science Fund
10.13039/501100002428
P 29192
Regularization Graphs for Variational Imaging
FWF Austrian Science Fund
10.13039/501100002428
W 1244
Partial Differential Equations - Modelling, Analysis, Numerical Methods and Optimization