pyALDVC: A Python Implementation of Augmented Lagrangian Digital Volume Correlation

3-D displacement and strain, inside the material

Open-source digital volume correlation for micro-CT, confocal and other 3-D scans:
a desktop application, a command line and a Python library, with optional GPU acceleration.

Website: zachtong.github.io/pyALDVC
_{cases, how it works, accuracy and speed, all in one page}

_{Sphere indentation of a hydrogel, confocal scan of 1024 × 1024 × 306 voxels: vertical displacement on the deformed node grid, down to −10.5 voxels (−4.5 µm) under the sphere; 144,342 nodes in 71 s on one GPU. Measured with pyALDVC (local subset solver).}

pyALDVC measures full-field displacement and strain inside a material from a sequence of 3-D scans. It is the Python version of the MATLAB ALDVC code (Yang, Hazlett, Landauer, Franck, Exp. Mech. 2020) and the volumetric sibling of pyALDIC: free, open source, and usable without writing a line of code.

Cases

Foam under compression Micro-CT slices, 987 × 1009 × 1856 voxels	Foam: displacement on the node grid 48,720 nodes in about 3 min; deformation exaggerated 2×
Rigid-body rotation Synthetic, six 5° steps to 30° (looped forward and back); tracked in 34 s	Laser-induced cavitation Residual von Mises strain (unpublished data)

The full cases, with the experiments behind them, on the website →

_{The foam, hydrogel and rotation fields were measured with pyALDVC (local subset solver) on one GPU. Their data are part of the DVC Challenge 2.0 dataset, doi:10.18434/mds2-4129, described in Tong, Z. et al. Digital Volume Correlation Challenge 2.0: A Comprehensive Dataset for Digital Volume Correlation Benchmarking. Research Square preprint (2026). https://doi.org/10.21203/rs.3.rs-9683321/v1. Foam data courtesy of NIST (Landauer et al., Sci. Data 10, 356, 2023).}

What it does

How it works

The method, and the tools that tell you how far to trust a result. Every figure is computed from synthetic volumes with a known answer, by a script in this repository. The website explains each one in full, with the AL-DVC method itself.

Tracking a sequence. Accumulative tracking (the default) correlates every scan with the first, so errors do not add up; incremental tracking correlates each scan with the one before and chains the steps, and follows motion too large for one step.

Texture analysis. How far the grey values stay correlated sets the subset size: four correlation lengths per axis, past which a larger subset barely lowers the error.

How precise is the result? A noise floor from static scans, the predicted error of every node, and confidence intervals that account for correlated neighbours.

Removing rigid-body motion. A specimen that shifts or turns adds displacement that is not deformation, and a rotation reads as false strain. A closed-form rigid fit removes it and reveals the deformation underneath.

Accuracy and speed

_{Synthetic rows: RMS error of each displacement component at the interior nodes, subset 16, step 8 voxels, default settings. Confocal rows: subset 32, step 8 voxels, with the settings of the MATLAB example run. Tested on Python 3.10, 3.11 and 3.12 on every push to main.}

Install

conda create -n pyaldvc python=3.12 -y
conda activate pyaldvc
pip install al-dvc
al-dvc

pip install "al-dvc[gpu]" in place of the third line adds NVIDIA GPU support; it needs the NVIDIA driver, not the CUDA Toolkit. The last command opens the application, and al-dvc --self-test checks the install. Optional: nibabel, pynrrd and pydicom read NIfTI, NRRD and DICOM; imageio with imageio-ffmpeg records MP4.

No Python? Every release ships a portable Windows bundle (CPU only; for NVIDIA GPU acceleration, install with pip as above): unzip it and double-click pyALDVC.exe. Then read the user guide.

Citation

If pyALDVC helps your research, please cite the software (the concept DOI always resolves to the latest release) and the method:

Tong, Z., Yang, J. pyALDVC: Augmented Lagrangian Digital Volume Correlation in Python. Zenodo (2026). https://doi.org/10.5281/zenodo.22883767

Yang, J., Hazlett, L., Landauer, A. K., Franck, C. Augmented Lagrangian Digital Volume Correlation (ALDVC). Experimental Mechanics 60, 1205–1223 (2020). https://doi.org/10.1007/s11340-020-00607-3

@software{tong_pyaldvc,
  author    = {Tong, Zixiang and Yang, Jin},
  title     = {{pyALDVC}: Augmented Lagrangian Digital Volume Correlation in Python},
  publisher = {Zenodo},
  year      = {2026},
  doi       = {10.5281/zenodo.22883767},
  url       = {https://github.com/zachtong/pyALDVC}
}

@article{yang_aldvc_2020,
  author  = {Yang, Jin and Hazlett, Lauren and Landauer, Alexander K. and Franck, Christian},
  title   = {Augmented {Lagrangian} Digital Volume Correlation ({ALDVC})},
  journal = {Experimental Mechanics},
  volume  = {60},
  number  = {9},
  pages   = {1205--1223},
  year    = {2020},
  doi     = {10.1007/s11340-020-00607-3}
}

Licence

BSD 3-Clause. Developed by Zixiang Tong and Jin Yang in Dr. Jin Yang's group at The University of Texas at Austin.