Datasets

This dataset contains two experimentally measured X-ray scans, one reference scan and one scan in which the object translates while rotating. The reference scan consists of 3600 projection images, with one flat field and dark field image. The roto-translational scan consists of 360 projections, also with a flat field and dark field image. The acquisition files containing all relevant specifications of the scanner and the acquisition settings are also supplied. The code for reconstruction is available at GitHub.

Code for our publication "Designing Interpretable Recurrent Neural Networks for Video Reconstruction via Deep Unfolding". >> Code maintained on https://gitlab.com/etrovub/mlsp/reweighted-rnn << Citation: H. V. Luong, B. Joukovsky and N. Deligiannis, "Designing Interpretable Recurrent Neural Networks for Video Reconstruction via Deep Unfolding," in IEEE Transactions on Image Processing, vol. 30, pp. 4099-4113, 2021, doi: 10.1109/TIP.2021.3069296.

4D-CT lab-based experiment of hygroscopically swelling MDF over 5h+, imaged at 2.5um. MCR.zip: motion-compensated reconstructions of all scans. Please contact us if you would like the raw projections instead (~70GB) DVC_CNN_results.zip: vertical strain fields computed between the initial scan and scan i, with either DVC software. The CNN results refer to the binary image found in either folder: the MDF structure was segmented between fiber bundles (1 value) and the rest of the material (0 value). manuscript_figures_processed_data: processed data is directly used to generate Latex figures. The ...

Many environmental and industrial processes depend on how fluids displace each other in porous materials. However, the flow dynamics that govern this process are still poorly understood. This dataset introduces a novel 4D micro-velocimetry method to study this, based on tracking flow-tracing micro-particles (silver-coated hollow glass spheres of 10 µm diameter) using high-resolution X-ray computed tomography with fast imaging rates. The measurement was done at the TOMCAT beamline of the Swiss Light Source. The dataset pertains to unsteady-state multiphase flow in a porous sintered glass ...

Using synchrotron computed tomography, we have obtained 3D images of a carbon fiber-epoxy composite laminate loaded under tension. A double-edge notched specimen with a notch section area of ~ 1 mm × 1 mm is cut using waterjet machining from the [904/05]s laminate with 7-µm nominal fiber diameter. It was loaded under continuous tension, with a deformation rate of ~ 1.4 μm/s, and in-situ scanned with a propagation distance of 60 mm, a beam energy of 20 kV, and 1000 projections per scan, each with an exposure time of 9 ms. The magnification was 10×, and the voxel size was 1.1 μm. The test was ...

This data package contains the reconstructed scans presented in the paper of Beyer et al. 2024, Sulfide melt wetting properties in Earth's mantle: New constraints from combined 2D and 3D imaging. The data package is divided in five folders, each representing one sample. The goal of the study was to gain a better understanding of how polymineralic silicate rocks and pressure influence the mobility of sulfide melts in the Earth's deeper upper mantle. A series of high-pressure experiments was conducted between 6 and 13 GPa, with temperatures above the liquidus of FeNiS and a metal-to-S ratio of ...

The data sets and the processing scripts are a part of the publication: Event driven 4D STEM acquisition with a Timepix3 detector: microsecond dwelltime and faster scans for high precision and low dose applications (arXiv:2107.02864). The programming language used to process the data is written with python 3. In this work, a Timepix3 detector is used to perform 4D STEM experiments where instead of a frame based detection, a event driven approach is used. The experiments where performed on a probe-corrected FEI Themis Z with an acceleration voltage of 60 and 200 kV. The files provided show how ...

Even at low temperatures, metal nanoparticles (NPs) possess atomic dynamics that are key for their properties but challenging to elucidate. Recent experimental advances allow obtaining atomic‐resolution snapshots of the NPs in realistic regimes, but data acquisition limitations hinder the experimental reconstruction of the atomic dynamics present within them. Molecular simulations have the advantage that these allow directly tracking the motion of atoms over time. However, these typically start from ideal/perfect NP structures and, suffering from sampling limits, provide results that are ...

This repository contains [i] all images used to determine the onset of nucleation (data points plotted in the linked paper), [ii] all files (e.g. 3D models) required for reconstruction of the custom-built experimental setup, and [iii] all purposely written coding to operate this experimental setup. The aim of providing this data is to give more insight into the applied experimental procedures, one of the key messages of the linked paper.

(!) Because of the large size of the data repository (27GB), it cannot be downloaded as a single zip file. Multiple zip files are provided, each ...

This data package contains all the reconstructed µCT scans that were analyzed and displayed in the publication from Schröer et al. titled 3D Visualization of Cyanobacterial Biofilms using Micro-Computed Tomography with Contrast-Enhancing Staining Agents. The goal of this research and this data was to visualize cyanobacterial biofilms using micro-computed tomography using five different contras-enhancing staining agents (CESAs) Every folder contains a series of .tiff files making up a 3D volume and a text file that describes either the scanning or reconstruction parameters. The µCT scans show ...