Pix2Prof: fast extraction of sequential information from galaxy imagery via deep learning

  title={Pix2Prof: fast extraction of sequential information from galaxy imagery via deep learning},
  author={Michael J. Smith and Nikhil Arora and Connor Stone and St{\'e}phane Courteau and James E. Geach},
We present "Pix2Prof", a deep learning model that eliminates manual steps in the measurement of galaxy surface brightness (SB) profiles. We argue that a galaxy "profile" of any sort is conceptually similar to an image caption. This idea allows us to leverage image captioning methods from the field of natural language processing, and so we design Pix2Prof as a float sequence "captioning" model suitable for SB profile inferral. We demonstrate the technique by approximating the galaxy SB fitting… 

Figures and Tables from this paper

Galaxy Light Profile Convolutional Neural Networks (GaLNets). I. Fast and Accurate Structural Parameters for Billion-galaxy Samples

Next-generation large sky surveys will observe up to billions of galaxies for which basic structural parameters are needed to study their evolution. This is a challenging task that, for ground-based

AutoProf – I. An automated non-parametric light profile pipeline for modern galaxy surveys

We present an automated non-parametric light profile extraction pipeline called autoprof. All steps for extracting surface brightness (SB) profiles are included in autoprof, allowing streamlined

Predicting the Spectrum of UGC 2885, Rubin’s Galaxy with Machine Learning

Wu & Peek predict SDSS-quality spectra based on Pan-STARRS broadband grizy images using machine learning (ML). In this article, we test their prediction for a unique object, UGC 2885 (“Rubin’s

MaNGA galaxy properties – I. An extensive optical, mid-infrared photometric, and environmental catalogue

We present an extensive catalog of non-parametric structural properties derived from optical and mid-infrared imaging for 4585 galaxies from the MaNGA survey. DESI and WISE imaging are used to

The Intrinsic Scatter of Galaxy Scaling Relations

We present a compendium of disk galaxy scaling relations and a detailed characterization of their intrinsic scatter. Observed scaling relations are typically characterized by their slope, intercept,



Deep r-Band Photometry for Northern Spiral Galaxies

AstroVaDEr: astronomical variational deep embedder for unsupervised morphological classification of galaxies and synthetic image generation

AstroVaDEr demonstrates the ability to classify the morphology of galaxies in a human-interpretable way without the need for labelled training data, and the generative properties of the model allow for realistic synthetic images of galaxies to be sampled from the learned classification scheme.

A physically motivated definition for the size of galaxies in an era of ultradeep imaging

Present-day multi-wavelength deep imaging surveys allow to characterise the outskirts of galaxies with unprecedented precision. Taking advantage of this situation, we define a new physically

The 16th Data Release of the Sloan Digital Sky Surveys: First Release from the APOGEE-2 Southern Survey and Full Release of eBOSS Spectra

This paper documents the 16th data release (DR16) from the Sloan Digital Sky Surveys (SDSS), the fourth and penultimate from the fourth phase (SDSS-IV). This is the first release of data from the

Energy and Policy Considerations for Deep Learning in NLP

This paper quantifies the approximate financial and environmental costs of training a variety of recently successful neural network models for NLP and proposes actionable recommendations to reduce costs and improve equity in NLP research and practice.

Second data release of the Hyper Suprime-Cam Subaru Strategic Program

This paper presents the second data release of the Hyper Suprime-Cam Subaru Strategic Program, a wide-field optical imaging survey on the 8.2 meter Subaru Telescope, including a major update to the processing pipeline, including improved sky subtraction, PSF modeling, object detection, and artifact rejection.

Color and metallicity gradients in E and S0 galaxies

Photometric maps of the galaxies NGC 3115 (S0), NGC 3377 (E5), NGC 3379 (E0), and NGC 4762 (S0) are presented for four filter bandpasses: U, B, V, and K. In each of the galaxies, the color indices

The Properties and Luminosity Function of Extremely Low Luminosity Galaxies

We examine a sample of low-redshift (10 h-1 Mpc < d < 150 h-1 Mpc) field galaxies including galaxies with luminosities as low as Mr - 5 log h ~ -12.5, selected from the Sloan Digital Sky Survey Data

A Comprehensive Survey of Deep Learning for Image Captioning

This survey article presents a comprehensive review of existing deep-learning-based image captioning techniques and discusses the foundation of the techniques to analyze their performances, strengths, and limitations.