Galaxy Spectroscopy without Spectra: Galaxy Properties from Photometric Images with Conditional Diffusion Models

Modern spectroscopic surveys can only target a small fraction of the vast amount of photometrically cataloged sources in wide-field surveys. Here, we report the development of a generative artificial intelligence (AI) method capable of predicting optical galaxy spectra from photometric broadband ima...

Full description

Saved in:
Bibliographic Details
Main Authors: Lars Doorenbos, Eva Sextl, Kevin Heng, Stefano Cavuoti, Massimo Brescia, Olena Torbaniuk, Giuseppe Longo, Raphael Sznitman, Pablo Márquez-Neila
Format: Article
Language:English
Published: IOP Publishing 2024-01-01
Series:The Astrophysical Journal
Subjects:
Online Access:https://doi.org/10.3847/1538-4357/ad8bbe
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850138382303232000
author Lars Doorenbos
Eva Sextl
Kevin Heng
Stefano Cavuoti
Massimo Brescia
Olena Torbaniuk
Giuseppe Longo
Raphael Sznitman
Pablo Márquez-Neila
author_facet Lars Doorenbos
Eva Sextl
Kevin Heng
Stefano Cavuoti
Massimo Brescia
Olena Torbaniuk
Giuseppe Longo
Raphael Sznitman
Pablo Márquez-Neila
author_sort Lars Doorenbos
collection DOAJ
description Modern spectroscopic surveys can only target a small fraction of the vast amount of photometrically cataloged sources in wide-field surveys. Here, we report the development of a generative artificial intelligence (AI) method capable of predicting optical galaxy spectra from photometric broadband images alone. This method draws from the latest advances in diffusion models in combination with contrastive networks. We pass multiband galaxy images into the architecture to obtain optical spectra. From these, robust values for galaxy properties can be derived with any methods in the spectroscopic toolbox, such as standard population synthesis techniques and Lick indices. When trained and tested on 64 × 64 pixel images from the Sloan Digital Sky Survey, the global bimodality of star-forming and quiescent galaxies in photometric space is recovered, as well as a mass–metallicity relation of star-forming galaxies. The comparison between the observed and the artificially created spectra shows good agreement in overall metallicity, age, Dn4000, stellar velocity dispersion, and E ( B − V ) values. Photometric redshift estimates of our generative algorithm can compete with other current, specialized deep learning techniques. Moreover, this work is the first attempt in the literature to infer velocity dispersion from photometric images. Additionally, we can predict the presence of an active galactic nucleus up to an accuracy of 82%. With our method, scientifically interesting galaxy properties, normally requiring spectroscopic inputs, can be obtained in future data sets from large-scale photometric surveys alone. The spectra prediction via AI can further assist in creating realistic mock catalogs.
format Article
id doaj-art-14ffb6c4ec7f47e69b95d42182755f24
institution OA Journals
issn 1538-4357
language English
publishDate 2024-01-01
publisher IOP Publishing
record_format Article
series The Astrophysical Journal
spelling doaj-art-14ffb6c4ec7f47e69b95d42182755f242025-08-20T02:30:35ZengIOP PublishingThe Astrophysical Journal1538-43572024-01-01977113110.3847/1538-4357/ad8bbeGalaxy Spectroscopy without Spectra: Galaxy Properties from Photometric Images with Conditional Diffusion ModelsLars Doorenbos0https://orcid.org/0000-0002-0231-9950Eva Sextl1https://orcid.org/0009-0001-5618-4326Kevin Heng2https://orcid.org/0000-0003-1907-5910Stefano Cavuoti3https://orcid.org/0000-0002-3787-4196Massimo Brescia4https://orcid.org/0000-0001-9506-5680Olena Torbaniuk5https://orcid.org/0000-0003-4465-2564Giuseppe Longo6https://orcid.org/0000-0002-9182-8414Raphael Sznitman7https://orcid.org/0000-0001-6791-4753Pablo Márquez-Neila8https://orcid.org/0000-0001-5722-7618AIMI, ARTORG Center, University of Bern , Murtenstr. 50, CH-3008 Bern, Switzerland ; lars.doorenbos@unibe.chUniversitäts-Sternwarte , Fakultät für Physik, Ludwig-Maximilians Universität München, Scheinerstr. 1, 81679 München, Germany ; sextl@usm.lmu.deUniversitäts-Sternwarte , Fakultät für Physik, Ludwig-Maximilians Universität München, Scheinerstr. 1, 81679 München, Germany ; sextl@usm.lmu.deINAF—Astronomical Observatory of Capodimonte , Salita Moiariello 16, I-80131 Napoli, Italy; INFN—Sezione di Napoli , via Cinthia 9, 80126 Napoli, ItalyINAF—Astronomical Observatory of Capodimonte , Salita Moiariello 16, I-80131 Napoli, Italy; INFN—Sezione di Napoli , via Cinthia 9, 80126 Napoli, Italy; Department of Physics, University Federico II , Strada Vicinale Cupa Cintia, 21, 80126 Napoli, ItalyDepartment of Physics and Astronomy “Augusto Righi,” University of Bologna , via Piero Gobetti 93/2, 40129 Bologna, ItalyDepartment of Physics, University Federico II , Strada Vicinale Cupa Cintia, 21, 80126 Napoli, ItalyAIMI, ARTORG Center, University of Bern , Murtenstr. 50, CH-3008 Bern, Switzerland ; lars.doorenbos@unibe.chAIMI, ARTORG Center, University of Bern , Murtenstr. 50, CH-3008 Bern, Switzerland ; lars.doorenbos@unibe.chModern spectroscopic surveys can only target a small fraction of the vast amount of photometrically cataloged sources in wide-field surveys. Here, we report the development of a generative artificial intelligence (AI) method capable of predicting optical galaxy spectra from photometric broadband images alone. This method draws from the latest advances in diffusion models in combination with contrastive networks. We pass multiband galaxy images into the architecture to obtain optical spectra. From these, robust values for galaxy properties can be derived with any methods in the spectroscopic toolbox, such as standard population synthesis techniques and Lick indices. When trained and tested on 64 × 64 pixel images from the Sloan Digital Sky Survey, the global bimodality of star-forming and quiescent galaxies in photometric space is recovered, as well as a mass–metallicity relation of star-forming galaxies. The comparison between the observed and the artificially created spectra shows good agreement in overall metallicity, age, Dn4000, stellar velocity dispersion, and E ( B − V ) values. Photometric redshift estimates of our generative algorithm can compete with other current, specialized deep learning techniques. Moreover, this work is the first attempt in the literature to infer velocity dispersion from photometric images. Additionally, we can predict the presence of an active galactic nucleus up to an accuracy of 82%. With our method, scientifically interesting galaxy properties, normally requiring spectroscopic inputs, can be obtained in future data sets from large-scale photometric surveys alone. The spectra prediction via AI can further assist in creating realistic mock catalogs.https://doi.org/10.3847/1538-4357/ad8bbeGalaxy propertiesGalaxy photometryGalaxy spectroscopySky surveysNeural networksConvolutional neural networks
spellingShingle Lars Doorenbos
Eva Sextl
Kevin Heng
Stefano Cavuoti
Massimo Brescia
Olena Torbaniuk
Giuseppe Longo
Raphael Sznitman
Pablo Márquez-Neila
Galaxy Spectroscopy without Spectra: Galaxy Properties from Photometric Images with Conditional Diffusion Models
The Astrophysical Journal
Galaxy properties
Galaxy photometry
Galaxy spectroscopy
Sky surveys
Neural networks
Convolutional neural networks
title Galaxy Spectroscopy without Spectra: Galaxy Properties from Photometric Images with Conditional Diffusion Models
title_full Galaxy Spectroscopy without Spectra: Galaxy Properties from Photometric Images with Conditional Diffusion Models
title_fullStr Galaxy Spectroscopy without Spectra: Galaxy Properties from Photometric Images with Conditional Diffusion Models
title_full_unstemmed Galaxy Spectroscopy without Spectra: Galaxy Properties from Photometric Images with Conditional Diffusion Models
title_short Galaxy Spectroscopy without Spectra: Galaxy Properties from Photometric Images with Conditional Diffusion Models
title_sort galaxy spectroscopy without spectra galaxy properties from photometric images with conditional diffusion models
topic Galaxy properties
Galaxy photometry
Galaxy spectroscopy
Sky surveys
Neural networks
Convolutional neural networks
url https://doi.org/10.3847/1538-4357/ad8bbe
work_keys_str_mv AT larsdoorenbos galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels
AT evasextl galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels
AT kevinheng galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels
AT stefanocavuoti galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels
AT massimobrescia galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels
AT olenatorbaniuk galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels
AT giuseppelongo galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels
AT raphaelsznitman galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels
AT pablomarquezneila galaxyspectroscopywithoutspectragalaxypropertiesfromphotometricimageswithconditionaldiffusionmodels