Unpaired fast- and slow-acquisition microCT scans of carbon-fibre-reinforced composites

data.zip contains tomographic images of carbon-fibre-reinforced composites with Toray T700S fibres in a Sicomin SR8500 KTA313 epoxy matrix.

The datasets accompany the paper Deep-learning image enhancement and fibre segmentation from time-resolved computed tomography of fibre-reinforced composites.

They are used in a two-step process:1) to train a 3DCycleGAN enhancement algorithm available on https://github.com/pvilla/3DCycleGaN.2) to train a UNetID segmentation algorithm available on https://github.com/CMG-KULeuven/UnetID-FibreSeg.

The tomograms were acquired at the TOMCAT Beamline at the Swiss Light Source in October 2021.

Acquisition specifications:

T700_GF_0p8um_1ms:

detector: GigaFRoST

effective voxel size: 800x800x800 nm³

# projections: 500

exposure time: 1 ms

total acquisition time: 500 x 1 ms = 0.5 s

beam: polychromatic

datasets: (same sample, different regions)T700-T-21_GF_0p8um_1ms_1.h5T700-T-21_GF_0p8um_1ms_2.h5T700-T-21_GF_0p8um_1ms_3.h5

T700_GF_1p6um_0p5ms:

detector: GigaFRoST

effective voxel size: 1600x1600x1600 nm³

# projections: 500

exposure time: 0.5 ms

total acquisition time: 500 x 0.5 ms = 0.25 s

beam: polychromatic

files: (same sample, different regions)T700-T-21_GF_1p6um_0p5ms_1.h5T700-T-21_GF_1p6um_0p5ms_2.h5

T700_GF_1p6um_3ms:

detector: GigaFRoST

effective voxel size: 1600x1600x1600 nm³

# projections: 500

exposure time: 3 ms

total acquisition time: 500 x 3 ms = 1.5 s

beam: polychromatic

files: (same sample, different regions)T700-T-21_GF_1p6um_3ms_1.h5T700-T-21_GF_1p6um_3ms_2.h5T700-T-21_GF_1p6um_3ms_3.h5T700-T-21_GF_1p6um_3ms_4.h5T700-T-21_GF_1p6um_3ms_5.h5

T700_pco_0p4um:

detector: pco.edge 5.5

effective voxel size: 400x400x400 nm³ (scaled from 325x325x325 nm³)

# projections: 2000

exposure time: 250 ms

total acquisition time: 2000 x 250 ms = 500 s

beam: 15 keV

files: (different samples)T700-T-02_pco_0p4um.h5T700-T-08_pco_0p4um.h5T700-T-21_pco_0p4um_reference.h5T700-T-26_pco_0p4um.h5

3DCycleGAN_trained_models.zip contains pre-trained models for the 3DCycleGAN enhancement algorithm.

UnetID_trained_models.zip contains pre-trained models for the UNetID segmentation algorithm.