Deep learning approach for identification of HII regions during reionization in 21-cm observations -- II. foreground contamination

Abstract

The upcoming Square Kilometre Array Observatory (SKAO) will produce images of neutral hydrogen distribution during the epoch of reionization by observing the corresponding 21-cm signal. However, the 21-cm signal will be subject to instrumental limitations such as noise and galactic foreground contamination which pose a challenge for accurate detection. In this study, we present the SegU-Net v2 framework, an enhanced version of our convolutional neural network, built to identify neutral and ionized regions in the 21-cm signal contaminated with foreground emission. We trained our neural network on 21-cm image data processed by a foreground removal method based on Principal Component Analysis achieving an average classification accuracy of 71 per cent between redshift $z=7$ to $11$. We tested SegU-Net v2 against various foreground removal methods, including Gaussian Process Regression, Polynomial Fitting, and Foreground-Wedge Removal. Results show comparable performance, highlighting SegU-Net v2's independence on these pre-processing methods. Statistical analysis shows that a perfect classification score with $AUC=95\%$ is possible for $8<z<10$. While the network prediction lacks the ability to correctly identify ionized regions at higher redshift and differentiate well the few remaining neutral regions at lower redshift due to low contrast between 21-cm signal, noise and foreground residual in images. Moreover, as the photon sources driving reionization are expected to be located inside ionised regions, we show that SegU-Net v2 can be used to correctly identify and measure the volume of isolated bubbles with $V_{\mathrm{ion}}>(10\mathrm{cMpc}{)}^3$ at $z>9$, for follow-up studies with infrared/optical telescopes to detect these sources.

19 pages, 10 figures, 3 tables, 2 appendixes

References in corpus (41)

• Array Programming with NumPy
• scikit-image: Image processing in Python
• Five-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Cosmological Interpretation
• Cosmology at Low Frequencies: The 21 cm Transition and the High-Redshift Universe
• WSClean: an implementation of a fast, generic wide-field imager for radio astronomy
• Efficient Simulations of Early Structure Formation and Reionization
• Two Remarkably Luminous Galaxy Candidates at $z\approx 10-12$ Revealed by JWST
• Early results from GLASS-JWST. III: Galaxy candidates at z$\sim$9-15
• Stress Testing $\Lambda$CDM with High-redshift Galaxy Candidates
• Improved upper limits on the 21-cm signal power spectrum of neutral hydrogen at $\mathbf{z}\approx 9.1$ from LOFAR
• 21 cm fluctuations from inhomogeneous X-ray heating before reionization
• Foreground simulations for the LOFAR - Epoch of Reionization Experiment
• Revisiting the radio interferometer measurement equation. I. A full-sky Jones formalism
• Deep multi-redshift limits on Epoch of Reionisation 21cm Power Spectra from Four Seasons of Murchison Widefield Array Observations
• The Epoch of Reionization Window: II. Statistical Methods for Foreground Wedge Reduction
• Blind foreground subtraction for intensity mapping experiments
• 21cmFAST v3: A Python-integrated C code forgenerating 3D realizations of the cosmic 21cm signal
• The Scale of the Problem : Recovering Images of Reionization with GMCA
• An Improved Method for 21cm Foreground Removal
• Constraining the intergalactic medium at $z\approx$ 9.1 using LOFAR Epoch of Reionization observations
• Non-parametric foreground subtraction for 21cm epoch of reionization experiments
• HI intensity mapping with MeerKAT: power spectrum detection in cross-correlation with WiggleZ galaxies
• The First Galaxies: Assembly of Disks and Prospects for Direct Detection
• Will point sources spoil 21 cm tomography?
• De-contamination of cosmological 21-cm maps
• Bubble size statistics during reionization from 21-cm tomography
• Did JWST observe imprints of axion miniclusters or primordial black holes?
• Imaging the redshifted 21-cm pattern around the first sources during the cosmic dawn using the SKA
• Deep ALMA redshift search of a z~12 GLASS-JWST galaxy candidate
• Constraining the state of the intergalactic medium during the Epoch of Reionization using MWA 21-cm signal observations
• Visibility based angular power spectrum estimation in low frequency radio interferometric observations
• Detecting ionized bubbles in redshifted 21 cm maps
• Adding Context to JWST Surveys with Current and Future 21cm Radio Observations
• Foreground removal for Square Kilometre Array observations of the Epoch of Reionization with the Correlated Component Analysis
• A Short Introduction to Reionization Physics
• Inferring the properties of the sources of reionization using the morphological spectra of the ionized regions
• BEoRN: A fast and flexible framework to simulate the epoch of reionisation and cosmic dawn
• Detecting the HI Power Spectrum in the Post-Reionization Universe with SKA-Low
• Persistent topology of the reionization bubble network. II: Evolution & Classification
• Suppressing variance in 21-cm signal simulations during reionization
• Warm dark matter constraints from the JWST

Cited by in corpus (13)

• Square Kilometre Array Science Data Challenge 3a: foreground removal for an EoR experiment
• Spectral modelling of Cygnus A between 110 and 250 MHz. Impact on the LOFAR 21-cm signal power spectrum
• Detecting galaxy-21-cm cross-correlation during reionization
• First upper limits on the 21-cm signal power spectrum of neutral hydrogen at $z=9.16$ from the LOFAR 3C196 field
• Deep learning approach for identification of HII regions during reionization in 21-cm observations -- III. image recovery
• Field-level Reconstruction from Foreground-Contaminated 21-cm Maps
• Reionisation time fields reconstruction from 21 cm signal maps
• The 21cm-galaxy cross-correlation: Realistic forecast for 21cm signal detection and reionisation constraints
• Reionization Parameter Inference from 3D Minkowski Functionals of the 21 cm Signals
• An Alcock-Paczynski Test on Reionization Bubbles for Cosmology
• Mapping neutral islands during end stages of reionization with photometric intergalactic medium tomography
• Exploring the Early Universe with Deep Learning
• Deep learning with hybrid frequency differencing and principal component analysis for 21-cm foreground and beam mitigation