Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit

By Kevin Schawinski, Ce Zhang, Hantian Zhang, Lucas Fowler, G. Santhanam. 2017

210 citations recorded by the source corpus.

Observations of astrophysical objects such as galaxies are limited by various sources of random and systematic noise from the sky background, the optical system of the telescope and the detector used to record the data. Conventional deconvolution techniques are limited in their ability to recover features in imaging data by the Shannon-Nyquist sampling theorem. Here we train a generative adversarial network (GAN) on a sample of $4,550$ images of nearby galaxies at $0.01<z<0.02$ from the Sloan Digital Sky Survey and conduct $10\times$ cross validation to evaluate the results. We present a method using a GAN trained on galaxy images that can recover features from artificially degraded images with worse seeing and higher noise than the original with a performance which far exceeds simple deconvolution. The ability to better recover detailed features such as galaxy morphology from low-signal-to-noise and low angular resolution imaging data significantly increases our ability to study existing data sets of astrophysical objects as well as future observations with observatories such as the Large Synoptic Sky Telescope (LSST) and the Hubble and James Webb space telescopes.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

In Generative Models

Citations in the corpus

Deep Fluids: A Generative Network for Parameterized Fluid Simulations (2018, methodology): “GANs have also been employed to generate [RLM ∗ 17] and deconvolve [SZZ ∗ 17] galaxy images, and reconstruct three-dimensional porous media [MDB17].”
Convolutional Neural Networks for Inverse Problems in Imaging: A Review (2017, background): “The same is true for deblurring, where training pairs can be generated by blurring [12]–[14].”
ContraGAN: Contrastive Learning for Conditional Image Generation (2020, background): “…can expand to various image processing applications and can learn the representations of high-dimensional datasets, scientists can enhance the quality of astronomical images [55, 56], design complex architectured materials [57], and efﬁciently search chemical space for developing materials [58].”
The ZipML Framework for Training Models with End-to-End Low Precision: The Cans, the Cannots, and a Little Bit of Deep Learning (2016, result): “Previously published works [49, 39] provide most results mentioned in this thesis.”