Hyperspectral X-ray CT Voxelized TV reconstruction of a single, iodine-stained lizard head sample

Dataset description

These datasets are voxel based reconstructions of hyperspectral CT data using the Core Imaging Library (CIL).  They are stored as NeXus files (derived from hdf5) which can be read in, visualised and manipulated using CIL.

 - PDHG_TV_1000_Sp_alpha_0.004.nxs

Is the solution after 1000 iterations of PDHG with TV applied in the spatial domain. 

 - PDHG_TV_1000_SpCh_alpha_0.003_beta_0.5.nxs

Is the solution after 1000 iterations of PDHG with TV applied both in the spatial domain, and in the energy (channel) domain. 

Dataset intended use

These datasets are used in the CIL training notebook:

https://github.com/TomographicImaging/CIL-Demos/blob/main/examples/3_Multichannel/03_Hyperspectral_reconstruction.ipynb

They can be imported using CIL, with the following code snippet:

from cil.io import NEXUSDataReader

reader = NEXUSDataReader(file_name='path/to/data/PDHG_TV_1000_Sp_alpha_0.004.nxs')

data = reader.read()