Skip to content
@UNSAT3D

UNSAT3D

U-Net segmentation of 3D micro-CT images of rooted soils using label data from multi-physics simulators

In a nutshell

Abstract

The health of soil is equally important as the health of humans; soil has to be strong enough to sustain the built infrastructures while being functional to allow water and nutrient transport to plants. While image processing has been a great help to support medical doctors in diagnosis, the use of images is not so straightforward for soil characterization. Difficulties arise simply because soil is random in nature (it contains e.g., solid grains, water, and plant roots) and has living organisms (e.g., plants) constantly modifying its structure. The project UNSAT tries to improve the identification capability of 3D image processing by training machine learning classifiers with high-fidelity, physics-based simulation data. We are interested in how new machine-learning techniques can help us to observe water transport and how root and soil react to water cycles during the growth of young maize.

Resources

  • References
    • 3D U-NET Learning dense volumetric segmentation from sparse annotation link.
    • Trainable Weka Segmentation: a machine learning tool for microscopy pixel classification link.
    • Hydro-micromechanical modeling of wave propagation in saturated granular crystals link.
    • Micro-scale investigation of unsaturated sand in mini-triaxal shearing using X-ray CT link.
    • Investigating the effect of porosity on the soil water retention curve using the multiphase Lattice Boltzmann Method link.
    • Root-soil interaction. Floriana Anselmucci's thesis: Root-soil interaction: effects on soil microstructure.
    • Traditional image processing vs deep learning link
  • Software
  • Datasets
    • X-ray Computed Tomography Reconstructions of Partially Saturated Vegetated Sand
      • The published selection: link
      • A bigger 30Gb selection on surfdrive
    • Phase field data generated from coupled Lattice Boltzmann-discrete element simulations link
    • Rooted soil observed in vivo in 4D through x-ray CT link
  • Learning materials

Other


Project logo: Soil icons created by Freepik - Flaticon

Popular repositories Loading

  1. XrayTIFF2h5 XrayTIFF2h5 Public

    Script for data collection

    Python 1

  2. unsat unsat Public

    Input/Output tools for the UNSAT project

    Python 1

  3. .github .github Public

    Project's homepage

  4. sandbox sandbox Public

    Initial exploration

    Jupyter Notebook

  5. Awesome-U-Net Awesome-U-Net Public

    Forked from NITR098/Awesome-U-Net

    Fork for using pretrained models

    Jupyter Notebook

  6. notes notes Public

Repositories

Showing 7 of 7 repositories
  • unsat Public

    Input/Output tools for the UNSAT project

    UNSAT3D/unsat’s past year of commit activity
    Python 1 Apache-2.0 0 8 0 Updated Oct 14, 2024
  • XrayTIFF2h5 Public

    Script for data collection

    UNSAT3D/XrayTIFF2h5’s past year of commit activity
    Python 1 Apache-2.0 0 0 0 Updated Aug 7, 2024
  • notes Public
    UNSAT3D/notes’s past year of commit activity
    0 0 0 0 Updated Nov 21, 2023
  • dino Public Forked from facebookresearch/dino

    PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO

    UNSAT3D/dino’s past year of commit activity
    Python 0 Apache-2.0 925 0 0 Updated Nov 17, 2023
  • .github Public

    Project's homepage

    UNSAT3D/.github’s past year of commit activity
    0 0 1 0 Updated Oct 27, 2023
  • Awesome-U-Net Public Forked from NITR098/Awesome-U-Net

    Fork for using pretrained models

    UNSAT3D/Awesome-U-Net’s past year of commit activity
    Jupyter Notebook 0 42 1 0 Updated Sep 27, 2023
  • sandbox Public

    Initial exploration

    UNSAT3D/sandbox’s past year of commit activity
    Jupyter Notebook 0 Apache-2.0 0 0 0 Updated Aug 21, 2023

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Top languages

Loading…

Most used topics

Loading…