OpenMBIR Software Developement Project:
Open Model-Based Iterative Reconstruction Software for Tomographic Reconstruction
Charles Bouman's web page
S. Venkat Venkatakrishnan's web page

OpenMBIR is a family of open source software development projects based on model-based iterative reconstruction (MBIR) algorithms for tomographic reconstruction. MBIR algorithms combine models of the physics of the sensor (i.e., forward model) with models of the image being reconstructed (i.e., prior model). By doing this, MBIR can form high quality reconstructions even with sparse and noisy data.

## Tomography

Super Voxel MBIR Python package - A python package for super-fast multi-core 3D parallel beam MBIR with an easy-to-use interface. This is what you will want if you are doing parallel beam CT for almost any application (i.e., TEM, X-ray microscopy, etc.).

OpenMBIR Index - Index to GitHub repositories for different OpenMBIR projects.

OpenMBIR-ParBeam - Reference code for parallel beam MBIR. This code is very slow, so it is really designed more as a reference implementation or for use as a code base for other implementations.

OpenMBIR-TEM - This is 3D MBIR code for electron microscopy. This code is robust with a easy to use GUI (see documentation below), but it does not use the fast super-voxel technology.
TEMBIR-HAADF - Tomography Environment for MBIR (TEMBIR) GUI and command line tool for HAADF-STEM tomography. This can handle the missing wedge, low-SNR and missing calibration parameters associated with the HAADF measurement.
TEMBIR-BF - Tomography Environment for MBIR (TEMBIR) GUI and command line tool for BF-(S)TEM tomography that can account for Bragg scatter. This can also be very useful for low-dose tomography.
TIMBIR X-ray - Time Interlaced Model-Based Iterative Reconstruction (TIMBIR). A command line tool for 3D and 4D Synchrotron X-ray micro-tomography reconstruction which also corrects for ring and streak artifacts. Code can run on a compute cluster.

Legacy code - This is older code that might still be useful, particularly for new code developement.

## Related Publications

paper Xiao Wang, Amit Sabne, Sherman Kisner, Anand Raghunathan, Charles Bouman, and Samuel Midkiff, High Performance Model Based Image Reconstruction,'' {\em 21st ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP '16),} March 12-16, 2016.

paper Amit Sabne, Xiao Wang, Sherman Kisner, Charles Bouman, Anand Raghunathan, and Samuel Midkiff, Model-based Iterative CT Image Reconstruction on GPUs,'' {\em 21st ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP '17),} February 4-8, 2017.

paper Xiao Wang, Amit Sabne, Putt Sakdhnagool, Sherman J. Kisner, Charles A. Bouman, and Samuel P. Midki, Massively Parallel 3D Image Reconstruction," 2017 Supercomputing Conference (SC17),'' ACM, November 13-16, 2017. (Selected as one of three finalists for 2017 ACM Gordon Bell Prize.)

paper S. V. Venkatakrishnan, et al. Making Advanced Scientific Algorithms and Big Scientific Data Management More Accessible,'' Electronic Imaging, Computational Imaging XIV, pp. 1-7(7), 2016.