Data

Tools

Indexes

Applications

Experiments

Open Source Science

auto-denoise

Unsupervised and self-supervised CNN denoising methods.

https://github.com/cea-metrocarac/auto-denoise

Science Score: 39.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ✓
    DOI references
    Found 12 DOI reference(s) in README
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (10.1%) to scientific vocabulary
Last synced: 10 months ago · JSON representation

Repository

Unsupervised and self-supervised CNN denoising methods.

Basic Info
Statistics
  • Stars: 3
  • Watchers: 1
  • Forks: 0
  • Open Issues: 1
  • Releases: 1
Created about 2 years ago · Last pushed about 1 year ago
Metadata Files
Readme Changelog Contributing Funding License Code of conduct Codemeta

README.md

Auto-Denoise

ci documentation pypi version gitter Code style: black

Auto-denoise (autoden) provides implementations for a small selection of unsupervised and self-supervised CNN denoising methods. These methods currently include:

  • Noise2Noise (N2N) - A self-supervised denoising method using pairs of images of the same object [1].
  • Noise2Void (N2V) - A self-supervised denoising method capable of working with a single image [2]. We have also implemented a later development of the method that can work with structured noise [3].
  • Deep Image Prior (DIP) - An unsupervised denoising/upsampling/deconvolution method that can also work with a single image [4].

We also provide example implementations of supervised denoising methods, and the tomography specific Noise2Inverse (N2I) method [5].

References:

  • [1] J. Lehtinen et al., “Noise2Noise: Learning Image Restoration without Clean Data,” in Proceedings of the 35th International Conference on Machine Learning, J. Dy and A. Krause, Eds., in Proceedings of Machine Learning Research, vol. 80. PMLR, 2018, pp. 2965–2974. https://proceedings.mlr.press/v80/lehtinen18a.html
  • [2] A. Krull, T.-O. Buchholz, and F. Jug, “Noise2Void - Learning Denoising From Single Noisy Images,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2019, pp. 2124–2132. doi: 10.1109/CVPR.2019.00223.
  • [3] C. Broaddus, A. Krull, M. Weigert, U. Schmidt, and G. Myers, “Removing Structured Noise with Self-Supervised Blind-Spot Networks,” in 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI), IEEE, Apr. 2020, pp. 159–163. doi: 10.1109/ISBI45749.2020.9098336.
  • [4] V. Lempitsky, A. Vedaldi, and D. Ulyanov, “Deep Image Prior,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, Jun. 2018, pp. 9446–9454. doi: 10.1109/CVPR.2018.00984.
  • [5] A. A. Hendriksen, D. M. Pelt, and K. J. Batenburg, "Noise2Inverse: Self-Supervised Deep Convolutional Denoising for Tomography," IEEE Transactions on Computational Imaging, vol. 6, pp. 1320–1335, 2020, doi: 10.1109/TCI.2020.3019647.

Getting Started

It takes just a few steps to setup Auto-Denoise on your machine.

Installing with conda

We recommend using Miniforge. Once installed miniforge, simply install autoden with: bash conda install auto-denoise -c n-vigano

Installing from PyPI

Simply install with: bash python -m pip install auto-denoise

If you are on jupyter, and don't have the rights to install packages system-wide, then you can install with: python ! python -m pip install --user auto-denoise

Installing from source

To install Auto-Denoise, simply clone this github.com project with either: bash git clone https://github.com/CEA-MetroCarac/auto-denoise.git auto-denoise or: bash git clone git@github.com:CEA-MetroCarac/auto-denoise.git auto-denoise

Then go to the cloned directory and run pip installer: bash cd auto-denoise pip install -e .

How to contribute

Contributions are always welcome. Please submit pull requests against the main branch.

If you have any issues, questions, or remarks, then please open an issue on github.com.

Owner

  • Name: CEA-MetroCarac
  • Login: CEA-MetroCarac
  • Kind: organization
  • Location: France

Metrology and Characterization activities at the French Alternative Energies and Atomic Energy Commission

CodeMeta (codemeta.json) 

{
  "@context": "https://w3id.org/codemeta/3.0",
  "type": "SoftwareSourceCode",
  "applicationCategory": "Image processing",
  "author": [
    {
      "id": "https://orcid.org/0000-0003-1704-5574",
      "type": "Person",
      "affiliation": {
        "type": "Organization",
        "name": "UGA, CEA-IRIG, MEM"
      },
      "email": "nicola.vigano@cea.fr",
      "familyName": "Vigano",
      "givenName": "Nicola"
    },
    {
      "type": "Role",
      "schema:author": "https://orcid.org/0000-0003-1704-5574",
      "roleName": "Main developer"
    }
  ],
  "codeRepository": "https://github.com/CEA-MetroCarac/auto-denoise",
  "datePublished": "2025-05-06",
  "description": "Auto-denoise (autoden) provides implementations for a small selection of unsupervised and self-supervised CNN denoising methods.",
  "keywords": "denoising; self-supervised; unsupervised; machine-learning",
  "license": "https://spdx.org/licenses/MIT",
  "name": "Auto-denoise",
  "operatingSystem": "Any",
  "programmingLanguage": "Python"
}

GitHub Events

Total
  • Release event: 1
  • Delete event: 3
  • Push event: 36
  • Pull request event: 6
  • Create event: 5
Last Year
  • Release event: 1
  • Delete event: 3
  • Push event: 36
  • Pull request event: 6
  • Create event: 5

Packages

  • Total packages: 1
  • Total downloads:
    • pypi 19 last-month
  • Total dependent packages: 0
  • Total dependent repositories: 0
  • Total versions: 2
  • Total maintainers: 1
pypi.org: auto-denoise

Unsupervised and self-supervised CNN denoising methods.

  • Versions: 2
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 19 Last month
Rankings
Dependent packages count: 9.2%
Average: 30.5%
Dependent repos count: 51.8%
Maintainers (1)
nicolavigano
Last synced: 10 months ago

Dependencies

.github/workflows/ci.yml actions
  • actions/cache v3 composite
  • actions/checkout v4 composite
  • actions/setup-python v4 composite
pyproject.toml pypi
  • autoflake >=1.4 develop
  • black * develop
  • duty >=0.6.0 develop
  • flake8 >=3.7.0 develop
  • flake8-bandit >=2.1.2 develop
  • flake8-black >=0.2.1 develop
  • flake8-bugbear >=20.11.1 develop
  • flake8-builtins >=1.5.3 develop
  • flake8-comprehensions >=3.3.1 develop
  • flake8-docstrings >=1.5.0 develop
  • flake8-string-format >=0.3.0 develop
  • flake8-tidy-imports >=4.2.1 develop
  • flake8-variables-names >=0.0.4 develop
  • git-changelog >=0.5.0 develop
  • httpx >=0.16.1 develop
  • isort * develop
  • jinja2-cli >=0.7.0 develop
  • mkdocs >=1.2.2 develop
  • mkdocs-macros-plugin >=0.5.0 develop
  • mkdocs-material >=6.2.7 develop
  • mkdocstrings >=0.16.2 develop
  • pep8-naming >=0.11.1 develop
  • pydocstyle >=6.1.1 develop
  • pytest * develop
  • toml >=0.10.2 develop
  • types-toml >=0.10.1 develop
  • imageio *
  • matplotlib *
  • numpy >=1.21
  • python >=3.10
  • scikit-image *
  • scipy *
  • tqdm *
environment.yaml pypi