Realistic galaxy image simulation via score-based generative models
Abstract
We show that a denoising diffusion probabilistic model (DDPM), a class of score-based generative model, can be used to produce realistic mock images that mimic observations of galaxies. Our method is tested with Dark Energy Spectroscopic Instrument (DESI) grz imaging of galaxies from the Photometry and Rotation curve OBservations from Extragalactic Surveys (PROBES) sample and galaxies selected from the Sloan Digital Sky Survey. Subjectively, the generated galaxies are highly realistic when compared with samples from the real data set. We quantify the similarity by borrowing from the deep generative learning literature, using the ‘Fréchet inception distance’ to test for subjective and morphological similarity. We also introduce the ‘synthetic galaxy distance’ metric to compare the emergent physical properties (such as total magnitude, colour, and half-light radius) of a ground truth parent and synthesized child data set. We argue that the DDPM approach produces sharper and more realistic images than other generative methods such as adversarial networks (with the downside of more costly inference), and could be used to produce large samples of synthetic observations tailored to a specific imaging survey. We demonstrate two potential uses of the DDPM: (1) accurate inpainting of occluded data, such as satellite trails, and (2) domain transfer, where new input images can be processed to mimic the properties of the DDPM training set. Here we ‘DESI-fy’ cartoon images as a proof of concept for domain transfer. Finally, we suggest potential applications for score-based approaches that could motivate further research on this topic within the astronomical community.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Jan 28, 2022
- Source ID
- 10.1093/mnras/stac130
Entities
People
- Connor Stone
- J. E. Geach
- Michael J. Smith
- Nikhil Arora
- R. A. JACKSON
- Stéphane Courteau
Organizations
- Argonne National Laboratory
- California Institute of Technology
- Chinese Academy of Sciences
- Division of Astronomical Sciences
- ETH Zurich
- Financiadora de Estudos e Projetos
- German Research Foundation
- Higher Education Funding Council for England
- Jet Propulsion Laboratory
- Lawrence Berkeley National Laboratory
- National Aeronautics and Space Administration
- National Center for Supercomputing Applications
- National Natural Science Foundation of China
- National Research Foundation of Korea
- National Science Foundation
- Natural Sciences and Engineering Research Council
- Office of Science
- Ohio State University
- Queen's University
- Royal Society
- SLAC National Accelerator Laboratory
- Science and Technology Facilities Council
- Spanish National Research Council
- Stanford University
- Texas A&M University
- United States Department of Energy
- University College London
- University of Chicago
- University of Edinburgh
- University of Hertfordshire
- University of Illinois Urbana–Champaign
- University of Michigan
- University of Nottingham
- University of Pennsylvania
- University of Portsmouth
- University of Sussex
- Yonsei University