BtB-MCR: Digital Material Reconstruction and Characterization

BtB-MCR is a repository dedicated to fast digital material reconstruction techniques. The repository will serve as a collection of already existing algorithms for digital microstructure reconstruction (MCR) while also serving as testing ground for novel implementations and novel algorithms, with the mission of increased robustness and faster computational speeds.

While the package itself is a work in progress, the current capabilities are as follows:

• Parallel implementation of Y-T algorithm
• QSGS reconstructions
• Voronoi Tesselations
• Random Sphere-Packing Method (and other shapes!)

I also plan on including characterizations to the package. The current scope (not implemented yet) is as follows:

• Correlation functions sped up by CPU and GPU implementations.
• Tortuosity calculations
• ED-cPSD phase-size analysis.

For a complete list of future plans and the current scope of the work, please check out the links below:

• The BtB-MCR code is available on GitHub.
• Please refer to the publications for additional detail:
  • Adam, A., Fang, H., & Li, X. (2024). Effective thermal conductivity estimation using a convolutional neural network and its application in topology optimization. Energy and AI, 15, 2666–5468. https://doi.org/10.1016/j.egyai.2023.100310

Once more methods are available, I will update this page and include a documentation.