Astronomical Image Reconstruction with Convolutional Neural Networks
MIT License
Astronomical Image Reconstruction with Convolutional Neural Networks
This repository contains the code used for the numerical experiments of the foillowing paper:
R. Flamary, "Astronomical image reconstruction with convolutional neural networks", European Conference on Signal Processing (EUSIPCO), 2017.
The code is under MIT Licence but please refer to and cite the above paper if you use it for academic purposes.
Note that the code is not yet available due to necessary cleanup but feel free to send me an email if you cannot wait for the GitHub version.
In order to run, the code requires the following Python modules:
You can also use the Tensorflow backend in keras but I used the theano backend to save the models provided int his repository.
If you have not already installed them you can install the dependencies with PIP usng the following command
$ pip install numpy scipy matplotlib keras theano astropy
Numerical experiments results are available in the res directory to reproduce all Figures and Table in the paper.
Numerical experiments can be reproduced in 2 steps:
To do that you should run the following scripts: