aringh / generalized-sinkhorn-and-tomography

Code for the article "Generalized Sinkhorn iterations for regularizing inverse problems using optimal mass transport" by J. Karlsson and A. Ringh

Entropy-regularized optimal transport

This repository contains the code for the article "Generalized Sinkhorn iterations for regularizing inverse problems using optimal mass transport" by J. Karlsson and A. Ringh. To appear in SIAM Journal on Imaging Sciences. arXiv:1612.02273.

Contents

The code contains the following

• Files containing the implementation of the entropy-regularized optimal transport and its proximal operator, based on generalized Sinkhorn iterations, and other utilities.
• Two scripts containing the two examples, and one script for post-processing some of the images for the hand-example.

Note that the two hand images used in the article do not belong to the authors and are therefore not included.

Installing and running the code

Clone the repository and install ODL (version 0.6.0) and ASTRA (version 1.8). This can be done, e.g., by using miniconda run the following commands to set up a new environment (essentially follow the odl installation instructions)

• $ conda create -c odlgroup -n my_env python=3.6 odl=0.6.0 matplotlib pytest scikit-image spyder
• $ source activate my_env
• $ conda install -c astra-toolbox astra-toolbox=1.8

After this, the scripts can be run using, e.g., spyder.

Contact

Axel Ringh, PhD student
Department of Mathematics, KTH Royal Institute of Technology, Stockholm, Sweden
[email protected]

Johan Karlsson, Associate Professor
Department of Mathematics, KTH Royal Institute of Technology, Stockholm, Sweden
[email protected]