Skip to main content
U.S. flag

An official website of the United States government

Return to search results

Dataset for Accelerating In Situ X-ray Tomography Using Sparse Projections and Deep Learning

Published by National Institute of Standards and Technology | National Institute of Standards and Technology | Catalog Last Checked: September 02, 2026 at 07:19 PM | Dataset Last Updated: June 24, 2025
Abstract (from manuscript): In situ X-ray tomography experiments provide powerful insight into material deformation and damage evolution under mechanical load. However, laboratory-based tomography setups often suffer from long acquisition times that hamper real-time observation. In this work, we demonstrate a deep-learning-based reconstruction method capable of high-quality volume reconstructions from sparse (10X fewer) projection datasets. We apply this method to in situ tensile experiments on additively manufactured Inconel 718 dog-bone specimens, significantly increasing our sampling density from only a few tomography scans per experiment to over 20 scans at different load steps. We compare conventional FDK reconstructions using 1001 projections with both (a) FDK using only 101 projections, and (b) our trained deep-reconstruction (DR) model with 101 projections. We show that the deep-reconstruction images closely match those from the full-projection FDK, enabling effective 3D segmentation and quantification of evolving porosity in the test samples. The data here demonstration our approach to increasing data throughput and sampling density for in situ studies. Data Abstract This dataset accompanies the publication "Accelerating In Situ X-ray Tomography Using Sparse Projections and Deep Learning." It contains reconstructed X-ray microscopy data and analysis workflows used to generate the results presented in the manuscript. The repository is organized as follows: Image Data: TIFF stacks representing both deep-learning-based and Feldkamp-Davis-Kress (FDK) reconstructions of in situ X-ray tomography measurements. These reconstructed volumes form the basis of all subsequent analysis. Analysis Workflows: P2_dummy_segment_final.ipynb: Jupyter notebook for segmentation and quantitative analysis of the reconstructed image stacks. This workflow produces VTK mesh files for visualization of segmented pore structures and saves corresponding PKL files containing porosity statistics. P2_dummy_visualization_final.ipynb: Jupyter notebook for generating the plots included in the publication. This notebook consumes the porosity statistics (PKL files) generated in the segmentation step and also requires the accompanying loadcell_final file containing the experimentally measured force–displacement data. Together, these resources provide both the raw image reconstructions and the complete analysis pipeline, from segmentation to visualization, enabling reproduction of the figures and quantitative results reported in the manuscript.

Resources

233 resources available

Find Related Datasets

data.gov

An official website of the GSA's Technology Transformation Services

Looking for U.S. government information and services?
Visit USA.gov