Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

A Conditional Autoencoder for Galaxy Photometric Parameter Estimation

Journal Article · · Publications of the Astronomical Society of the Pacific
 [1];  [2];  [3];  [4]
  1. Harvard University, Cambridge, MA (United States); OSTI
  2. Harvard-Smithsonian Center for Astrophysics, Cambridge, MA (United States)
  3. Harvard University, Cambridge, MA (United States); Harvard-Smithsonian Center for Astrophysics, Cambridge, MA (United States)
  4. Harvard University, Cambridge, MA (United States)

Astronomical photometric surveys routinely image billions of galaxies, and traditionally infer the parameters of a parametric model for each galaxy. This approach has served us well, but the computational expense of deriving a full posterior probability distribution function is a challenge for increasingly ambitious surveys. In this paper, we use deep learning methods to characterize galaxy images, training a conditional autoencoder on mock data. The autoencoder can reconstruct and denoise galaxy images via a latent space engineered to include semantically meaningful parameters, such as brightness, location, size, and shape. Our model recovers galaxy fluxes and shapes on mock data with a lower variance than the Hyper Suprime-Cam photometry pipeline, and returns reasonable answers even for inputs outside the range of its training data. When applied to data in the training range, the regression errors on all extracted parameters are nearly unbiased with a variance near the Cramr-Rao bound.

Research Organization:
Harvard University, Cambridge, MA (United States)
Sponsoring Organization:
USDOE Office of Science (SC); NASA ROSES; National Science Foundation (NSF)
Grant/Contract Number:
SC0013718
OSTI ID:
1979376
Journal Information:
Publications of the Astronomical Society of the Pacific, Journal Name: Publications of the Astronomical Society of the Pacific Journal Issue: 1034 Vol. 134; ISSN 0004-6280
Publisher:
Astronomical Society of the Pacific (ASP)Copyright Statement
Country of Publication:
United States
Language:
English

References (26)

GalSim: The modular galaxy image simulation toolkit journal April 2015
scarlet: Source separation in multi-band images by Constrained Matrix Factorization journal July 2018
The effects of seeing on Sérsic profiles - II. The Moffat PSF journal December 2001
A Concise Reference to (Projected) Sérsic R 1/ n Quantities, Including Concentration, Profile Slopes, Petrosian Indices, and Kron Magnitudes journal January 2005
DAOPHOT - A computer program for crowded-field stellar photometry journal March 1987
The Sloan Digital Sky Survey: Technical Summary journal September 2000
The Second Data Release of the Sloan Digital Sky Survey journal July 2004
The Two Micron All Sky Survey (2MASS) journal February 2006
Imfit: a Fast, Flexible new Program for Astronomical Image Fitting journal January 2015
BEYOND ELLIPSE(S): ACCURATELY MODELING THE ISOPHOTAL STRUCTURE OF GALAXIES WITH ISOFIT AND CMODEL journal September 2015
Cosmology with cosmic shear observations: a review journal July 2015
Deep learning for galaxy surface brightness profile fitting journal December 2017
Photometry of high-redshift blended galaxies using deep learning journal December 2019
The Hyper Suprime-Cam SSP Survey: Overview and survey design journal September 2017
The Hyper Suprime-Cam software pipeline journal October 2017
Characterization and photometric performance of the Hyper Suprime-Cam Software Pipeline journal December 2017
Loss Functions for Image Restoration With Neural Networks journal March 2017
Machine learning and image analysis for morphological galaxy classification journal March 2004
Reducing the Dimensionality of Data with Neural Networks journal July 2006
Weak Lensing for Precision Cosmology journal September 2018
SEP: Source Extractor as a library journal October 2016
WISE PHOTOMETRY FOR 400 MILLION SDSS SOURCES journal January 2016
Improved Point-source Detection in Crowded Fields Using Probabilistic Cataloging journal September 2017
Multiband Probabilistic Cataloging: A Joint Fitting Approach to Point-source Detection and Deblending journal March 2020
Photometric Biases in Modern Surveys journal March 2020
The DECam Plane Survey: Optical Photometry of Two Billion Objects in the Southern Galactic Plane journal February 2018

Similar Records

Anomaly detection in Hyper Suprime-Cam galaxy images with generative adversarial networks
Journal Article · Mon Sep 13 00:00:00 EDT 2021 · Monthly Notices of the Royal Astronomical Society · OSTI ID:1886969

Hierarchical Modeling and Statistical Calibration for Photometric Redshifts
Journal Article · Sat Aug 10 00:00:00 EDT 2019 · The Astrophysical Journal (Online) · OSTI ID:1562502