https://github.com/jakeret/tf_unet

Generic U-Net Tensorflow implementation for image segmentation
https://github.com/jakeret/tf_unet

deep-learning image-segmentation neural-network tensorflow

Last synced: 8 days ago
JSON representation

Generic U-Net Tensorflow implementation for image segmentation

• Host: GitHub
• URL: https://github.com/jakeret/tf_unet
• Owner: jakeret
• License: gpl-3.0
• Created: 2016-08-12T08:02:22.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2020-05-05T09:29:55.000Z (about 4 years ago)
• Last Synced: 2024-02-28T21:32:11.646Z (4 months ago)
• Topics: deep-learning, image-segmentation, neural-network, tensorflow
• Language: Python
• Homepage:
• Size: 4.7 MB
• Stars: 1,888
• Watchers: 67
• Forks: 754
• Open Issues: 90
• Metadata Files:
  • Readme: README.rst
  • Changelog: HISTORY.rst
  • Contributing: CONTRIBUTING.rst
  • License: LICENSE

Lists

• awesome-deeplearning-resources - Generic U-Net Tensorflow implementation for image segmentation

README

        
=============================

Tensorflow Unet

=============================

.. image:: https://readthedocs.org/projects/tf-unet/badge/?version=latest

	:target: http://tf-unet.readthedocs.io/en/latest/?badge=latest

	:alt: Documentation Status

		

.. image:: http://img.shields.io/badge/arXiv-1609.09077-orange.svg?style=flat

        :target: http://arxiv.org/abs/1609.09077

.. image:: https://img.shields.io/badge/ascl-1611.002-blue.svg?colorB=262255

        :target: http://ascl.net/1611.002

.. image:: https://mybinder.org/badge.svg

        :target: https://mybinder.org/v2/gh/jakeret/tf_unet/master?filepath=demo%2Fdemo_toy_problem.ipynb

.. warning::

    This project is discontinued in favour of a Tensorflow 2 compatible reimplementation of this project found under https://github.com/jakeret/unet

This is a generic **U-Net** implementation as proposed by `Ronneberger et al. `_ developed with **Tensorflow**. The code has been developed and used for `Radio Frequency Interference mitigation using deep convolutional neural networks `_ .

The network can be trained to perform image segmentation on arbitrary imaging data. Checkout the `Usage `_ section or the included Jupyter notebooks for a `toy problem `_ or the `Radio Frequency Interference mitigation `_ discussed in our paper.

The code is not tied to a specific segmentation such that it can be used in a toy problem to detect circles in a noisy image.

.. image:: https://raw.githubusercontent.com/jakeret/tf_unet/master/docs/toy_problem.png

   :alt: Segmentation of a toy problem.

   :align: center

To more complex application such as the detection of radio frequency interference (RFI) in radio astronomy.

.. image:: https://raw.githubusercontent.com/jakeret/tf_unet/master/docs/rfi.png

   :alt: Segmentation of RFI in radio data.

   :align: center

Or to detect galaxies and star in wide field imaging data.

.. image:: https://raw.githubusercontent.com/jakeret/tf_unet/master/docs/galaxies.png

   :alt: Segmentation of a galaxies.

   :align: center

As you use **tf_unet** for your exciting discoveries, please cite the paper that describes the package::

	@article{akeret2017radio,

	  title={Radio frequency interference mitigation using deep convolutional neural networks},

	  author={Akeret, Joel and Chang, Chihway and Lucchi, Aurelien and Refregier, Alexandre},

	  journal={Astronomy and Computing},

	  volume={18},

	  pages={35--39},

	  year={2017},

	  publisher={Elsevier}

	}