• Home
  • Search
  • Deciphering galaxy images using machine vision – combining variational autoencoder and principal component analysis for feature extraction
  • https://doi.org/10.1093/mnras/staf1926Copy DOI Icon

Deciphering galaxy images using machine vision – combining variational autoencoder and principal component analysis for feature extraction

Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

ABSTRACT Understanding the visual features contained in galaxy images is of utmost importance for interpreting their underlying structures. Here, we present a machine vision approach, combining a variational autoencoder (VAE) framework with principal component analysis (PCA), to decipher galaxy images. Using mock $gri$-band images from the EAGLE simulation, the VAE finds that around 35 features are needed to describe the images. Adding the PCA, we identify an optimal range of 10–12 features needed to capture 99.9 per cent of the variance in galaxy images. The exact optimal number varies with galaxy complexity: disc-dominated galaxies require 12 features, bulge-dominated galaxies need 9, and intermediate systems require 10–11 features. Correlations between extracted PCA features and structural measurements reveal that the VAE prioritizes galaxy size during reconstruction, with half-light radius strongly correlating with the highest-ranked principal components. Subsequent features capture morphology-dependent characteristics: disc-dominated galaxies emphasize size, asymmetry, and position angle; bulge-dominated systems focus on size, concentration, and axis ratio; while intermediate galaxies show enhanced attention to Sérsic index, indicating greater emphasis on accurately reproducing both disc and bulge structures. The PCA process significantly reduces the entanglement of the features compared to the raw VAE latent features, decreasing the correlations with the half-light radius and the Sérsic index from $14.5 \pm 1.0$ and $6.0 \pm 1.5$ features, respectively, to only $2.0 \pm 1.0$ components after PCA. Using Uniform Manifold Approximation and Projection, we construct 2D visualizations that preserve neighbourhood relationships from the high-dimensional feature space. This demonstrates that machine vision can effectively distinguish galaxy populations across different morphological types, including systems with atypical structures that may be overlooked by traditional classification methods, providing a data-driven complement to conventional structural measurements.

Similar Papers
  • Abstract

229. A Novel ‘One Health’ Approach to Understanding the Relationship of Antimicrobial Resistance Characteristics Among Humans, Bovines, and Canines

  • Dec 04, 2021
  • Open Forum Infectious Diseases
  • Laurel Legenza +7
  • Research Article
  • Citations141

Deep variational auto-encoders: A promising tool for dimensionality reduction and ball bearing elements fault diagnosis

  • Jul 20, 2018
  • Structural Health Monitoring
  • Gabriel San Martin +3
  • Research Article

Evaluating discrepancies in dimensionality reduction for time-series single-cell RNA-sequencing data.

  • May 01, 2025
  • Briefings in bioinformatics
  • Maren Hackenberg +3
  • Research Article
  • Citations38

Towards a new classification of galaxies: principal component analysis of CALIFA circular velocity curves

  • Apr 12, 2017
  • Monthly Notices of the Royal Astronomical Society
  • V Kalinova +18
  • Supplementary Content
  • Citations1

Nichtlineare Hauptkomponentenanalyse auf Basis neuronaler Netze

  • Jun 05, 2002
  • edoc Publication server (Humboldt University of Berlin)
  • Matthias Scholz
  • Dissertation

Mixtures of Variational Autoencoders for Cluster Analysis in Latent Space

  • Dec 11, 2024
  • Mashfiqul Chowdhury
  • Research Article
  • Citations5

Behavior of Linear and Nonlinear Dimensionality Reduction for Collective Variable Identification of Small Molecule Solution-Phase Reactions.

  • Feb 28, 2022
  • Journal of Chemical Theory and Computation
  • Hung M Le +6
  • Research Article

Using Variational Autoencoders with Machine Learning Algorithms in Cyber Security Applications

  • Aug 26, 2025
  • Artificial Intelligence and Applications
  • Thomas Taylor +2
  • Conference Article
  • Citations4

Enhancing Zero-Shot Many to Many Voice Conversion via Self-Attention VAE with Structurally Regularized Layers

  • Sep 01, 2022
  • Ziang Long +3
  • Research Article

Stratifying pulmonary hypertension severity in newborns from multi-view echocardiograms using variational autoencoders

  • Nov 01, 2025
  • European Heart Journal
  • Samuel Ruipérez-Campillo +6
  • PDF
  • Research Article
  • Citations25

Visualization Framework for High-Dimensional Spatio-Temporal Hydrological Gridded Datasets using Machine-Learning Techniques

  • Feb 21, 2020
  • Water
  • Abeer Mazher
  • Conference Article

LARGE DATABASE OF AIS VESSEL TRAFFIC FOR SMART DECISION-MAKING BASED ON AI IN SHIPPING COMPANIES

  • Nov 04, 2025
  • Ernest Czermański +6
  • Research Article
  • Citations4

Identification of Beef Odors under Different Storage Day and Processing Temperature Conditions Using an Odor Sensing System

  • Aug 29, 2024
  • Sensors (Basel, Switzerland)
  • Yuanchang Liu +4
  • Research Article

Adulteration Detection of Multi-Species Vegetable Oils in Camellia Oil Using SICRIT-HRMS and Machine Learning Methods

  • Jan 24, 2026
  • Foods
  • Mei Wang +6
  • PDF
  • Research Article
  • Citations30

Visualizing Single-Cell RNA-seq Data with Semisupervised Principal Component Analysis

  • Aug 12, 2020
  • International Journal of Molecular Sciences
  • Zhenqiu Liu
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.