HAADF-STEM tomography is widely used in the physical sciences for 3-D reconstructions because it is not effected by the dynamical diffraction effects seen in BF-ET. Since the data comes from a limited-tilt acquisition, analytical methods like FBP and iterative method like SIRT can result in reconstructions with noise and streak artifacts.
In this work, we propose an MBIR method for HAADF-STEM tomography that can result in significant improvements in reconstruction quality.
Further, it can also account for missing calibration parameters that are associated with the measurement.
For details see
S.V. Venkatakrishnan, L.F. Drummy, M. Jackson, M. De Graef, J.P. Simmons, and C.A. Bouman, "A Model Based Iterative Reconstruction Algorithm for High Angle Annular Dark Field Scanning Transmission Electron Microscope (HAADF-STEM) Tomography," IEEE Transactions on Image Processing, Vol 22, Issue 11 (Nov 2013).
C++ Based Software (with GUI)
- Automatic initialization of parameters
- Reconstruct the volume at various voxels sizes
- Multi-threaded code using Intel TBB
YouTube application tutorials:
Software supports aligned input files stored in
FEI mrc format
New [4/20/2014] : Added support for non-FEI MRC. See this very short
Software Source Code
All code is available from the Git Repository server. (Master branch)