Deep learning for galaxy surface brightness profile fitting

Abstract

Numerous ongoing and future large area surveys (e.g. DES, EUCLID, LSST, WFIRST), will increase by several orders of magnitude the volume of data that can be exploited for galaxy morphology studies. The full potential of these surveys can only be unlocked with the development of automated, fast and reliable analysis methods. In this paper we present DeepLeGATo, a new method for two-dimensional photometric galaxy profile modeling, based on convolutional neural networks. Our code is trained and validated on analytic profiles (HST/CANDELS F160W filter) and it is able to retrieve the full set of parameters of one- component Sérsic models: total magnitude, effective radius, Sérsic index, axis ratio. We show detailed comparisons between our code and GALFIT. On simulated data, our method is more accurate than GALFIT and 3000 time faster on GPU (50 times when run on the same CPU). On real data, DeepLeGATo trained on simulations behaves similarly to GALFIT on isolated galaxies. With a fast domain adaptation step made with the 0.1 - 0.8 per cent the size of the training set, our code is easily capable to reproduce the results obtained with GALFIT even on crowded regions. DeepLeGATo does not require any human intervention beyond the training step, rendering it much automated than traditional profiling methods. The development of this method for more complex models (two-component galaxies, variable PSF, dense sky regions) could constitute a fundamental tool in the era of big data in astronomy.

17 pages, 9 figures

References in corpus (23)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Deep Learning in Neural Networks: An Overview
• CANDELS: The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey
• CANDELS: The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey - The Hubble Space Telescope Observations, Imaging Data Products and Mosaics
• Galaxy Zoo : Morphologies derived from visual inspection of galaxies from the Sloan Digital Sky Survey
• 3D-HST+CANDELS: The Evolution of the Galaxy Size-Mass Distribution since $z=3$
• Theano: new features and speed improvements
• LSST Science Book, Version 2.0
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• Structural Parameters of Galaxies in CANDELS
• The Kilo-Degree Survey
• The massive end of the luminosity and stellar mass functions: Dependence on the fit to the light profile
• A catalog of visual-like morphologies in the 5 CANDELS fields using deep-learning
• Star-galaxy Classification Using Deep Convolutional Neural Networks
• GALAPAGOS: From Pixels to Parameters
• Galaxy And Mass Assembly (GAMA): ${\mathcal{M}}_{\star }-R_{\mathrm{e}}$ relations of $z=0$ bulges, discs and spheroids
• ProFit: Bayesian Profile Fitting of Galaxy Images
• The Millennium Galaxy Catalogue: The luminosity functions of bulges and discs and their implied stellar mass densities
• Galaxy And Mass Assembly (GAMA): ugrizYJHK Sérsic luminosity functions and the cosmic spectral energy distribution by Hubble type
• PyMorph: Automated Galaxy Structural Parameter Estimation using Python
• Galaxy And Mass Assembly (GAMA): Understanding the wavelength dependence of galaxy structure with bulge-disc decompositions
• Unsupervised feature-learning for galaxy SEDs with denoising autoencoders

Cited by in corpus (59)

• Galaxy Zoo: Probabilistic Morphology through Bayesian CNNs and Active Learning
• LinKS: Discovering galaxy-scale strong lenses in the Kilo-Degree Survey using Convolutional Neural Networks
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Morpheus: A Deep Learning Framework For Pixel-Level Analysis of Astronomical Image Data
• Transfer learning for galaxy morphology from one survey to another
• Deep Learning Identifies High-z Galaxies in a Central Blue Nugget Phase in a Characteristic Mass Range
• The evolution of galaxy shapes in CANDELS: from prolate to oblate
• Detecting outliers in astronomical images with deep generative networks
• Morphological classification of galaxies with deep learning: comparing 3-way and 4-way CNNs
• Galaxy Morphology Network: A Convolutional Neural Network Used to Study Morphology and Quenching in $\sim 100,000$ SDSS and $\sim 20,000$ CANDELS Galaxies
• Star formation rates and stellar masses from machine learning
• Testing Convolutional Neural Networks for finding strong gravitational lenses in KiDS
• AutoProf -- I. An automated non-parametric light profile pipeline for modern galaxy surveys
• Pushing automated morphological classifications to their limits with the Dark Energy Survey
• Machine-learning classification of astronomical sources: estimating F1-score in the absence of ground truth
• xGASS: The Impact of Photometric Bulges on the Scatter of HI Scaling Relations
• Discovery and analysis of low-surface-brightness galaxies in the environment of NGC 1052
• Quantifying Non-parametric Structure of High-redshift Galaxies with Deep Learning
• Weak-lensing shear measurement with machine learning: teaching artificial neural networks about feature noise
• Euclid preparation XXVI. The Euclid Morphology Challenge. Towards structural parameters for billions of galaxies
• A Method to Distinguish Quiescent and Dusty Star-forming Galaxies with Machine Learning
• Quasar microlensing light curve analysis using deep machine learning
• Euclid preparation. XXV. The Euclid Morphology Challenge -- Towards model-fitting photometry for billions of galaxies
• GAlaxy Light profile convolutional neural NETworks (GaLNets). I. fast and accurate structural parameters for billion galaxy samples
• PSFGAN: a generative adversarial network system for separating quasar point sources and host galaxy light
• Inferring galaxy dark halo properties from visible matter with Machine Learning
• A Robust Study of High-Redshift Galaxies: Unsupervised Machine Learning for Characterising morphology with JWST up to z ~ 8
• Euclid preparation: XXII. Selection of Quiescent Galaxies from Mock Photometry using Machine Learning
• Studying the Physical Properties of Tidal Features I. Extracting Morphological Substructure in CANDELS Observations and VELA Simulations
• Deep Learning-based galaxy image deconvolution
• MaNGA galaxy properties -- I. An extensive optical, mid-infrared photometric, and environmental catalogue
• Searching for hot subdwarf stars from the LAMOST spectra. III. classifying the hot subdwarf stars from LAMOST DR4 using deep learning method
• Deep learning for Sunyaev-Zel'dovich detection in Planck
• A Container-Based Workflow for Distributed Training of Deep Learning Algorithms in HPC Clusters
• Self-supervised Learning with Physics-aware Neural Networks I: Galaxy Model Fitting
• GaMPEN: A Machine Learning Framework for Estimating Bayesian Posteriors of Galaxy Morphological Parameters
• Morphological Parameters and Associated Uncertainties for 8 Million Galaxies in the Hyper Suprime-Cam Wide Survey
• HOLISMOKES -- XI. Evaluation of supervised neural networks for strong-lens searches in ground-based imaging surveys
• Classifying the formation processes of S0 galaxies using Convolutional Neural Networks
• Improving machine learning-derived photometric redshifts and physical property estimates using unlabelled observations
• Pix2Prof: fast extraction of sequential information from galaxy imagery via a deep natural language 'captioning' model
• HOLISMOKES XIII: Strong-lens candidates at all mass scales and their environments from the Hyper-Suprime Cam and deep learning
• Euclid preparation. LI. Forecasting the recovery of galaxy physical properties and their relations with template-fitting and machine-learning methods
• The Complex Structure of the Bulge of M31
• Representation learning for automated spectroscopic redshift estimation
• AGNet: Weighing Black Holes with Deep Learning
• GPU-accelerated Image Reduction Pipeline
• Deep learning-based deconvolution for interferometric radio transient reconstruction
• Predicting bulge to total luminosity ratio of galaxies using deep learning
• Anomaly detection in Astrophysics: a comparison between unsupervised Deep and Machine Learning on KiDS data
• Image feature extraction and galaxy classification: a novel and efficient approach with automated machine learning
• Using Machine Learning to Determine Morphologies of $z<1$ AGN Host Galaxies in the Hyper Suprime-Cam Wide Survey
• A new way to constrain the densities of intra-group medium in groups of galaxies with convolutional neural networks
• AGN -- host galaxy photometric decomposition using a fast, accurate and precise deep learning approach
• Euclid preparation. Estimating galaxy physical properties using CatBoost chained regressors with attention
• Neutron star envelopes with machine learning: a single-hidden-layer neural network application
• Automatic Machine Learning Framework to Study Morphological Parameters of AGN Host Galaxies within $z<1.4$ in the Hyper Supreme-Cam Wide Survey
• Non-Sequential Neural Network for Simultaneous, Consistent Classification and Photometric Redshifts of OTELO Galaxies
• ShapeNet: Shape Constraint for Galaxy Image Deconvolution