Deep learning-based super-resolution and de-noising for XMM-newton images

S.F. Sweere; I. Valtchanov; M. Lieu; A. Vojtekova; E. Verdugo; M. Santos-Lleo; F. Pacaud; A. Briassouli; D. Campora Perez

doi:10.1093/mnras/stac2437

Deep learning-based super-resolution and de-noising for XMM-newton images

S.F. Sweere^*, I. Valtchanov^*, M. Lieu, A. Vojtekova, E. Verdugo, M. Santos-Lleo, F. Pacaud, A. Briassouli, D. Campora Perez

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

The field of artificial intelligence based image enhancement has been rapidly evolving over the last few years and is able to produce impressive results on non-astronomical images. In this work, we present the first application of Machine Learning based super-resolution (SR) and de-noising (DN) to enhance X-ray images from the European Space Agency's XMM-Newton telescope. Using XMM-Newton images in band [0.5, 2] keV from the European Photon Imaging Camera pn detector (EPIC-pn), we develop XMM-SuperRes and XMM-DeNoise - deep learning-based models that can generate enhanced SR and DN images from real observations. The models are trained on realistic XMM-Newton simulations such that XMM-SuperRes will output images with two times smaller point-spread function and with improved noise characteristics. The XMM-DeNoise model is trained to produce images with 2.5× the input exposure time from 20 to 50 ks. When tested on real images, DN improves the image quality by 8.2 per cent, as quantified by the global peak-signal-to-noise ratio. These enhanced images allow identification of features that are otherwise hard or impossible to perceive in the original or in filtered/smoothed images with traditional methods. We demonstrate the feasibility of using our deep learning models to enhance XMM-Newton X-ray images to increase their scientific value in a way that could benefit the legacy of the XMM-Newton archive.

Original language	English
Pages (from-to)	4054-4069
Number of pages	16
Journal	Monthly Notices of the Royal Astronomical Society
Volume	517
Issue number	3
DOIs	https://doi.org/10.1093/mnras/stac2437
Publication status	Published - 1 Dec 2022

Keywords

techniques: high angular resolution
techniques: image processing
X-rays: general
PHOTON IMAGING CAMERA
ILLUSTRISTNG SIMULATIONS
CHANDRA
DECONVOLUTION
SIMILARITY
CLUSTERS

Access to Document

10.1093/mnras/stac2437Licence: CC BY

Cite this

@article{69083ed2999741408fb6c5566df80326,

title = "Deep learning-based super-resolution and de-noising for XMM-newton images",

abstract = "The field of artificial intelligence based image enhancement has been rapidly evolving over the last few years and is able to produce impressive results on non-astronomical images. In this work, we present the first application of Machine Learning based super-resolution (SR) and de-noising (DN) to enhance X-ray images from the European Space Agency's XMM-Newton telescope. Using XMM-Newton images in band [0.5, 2] keV from the European Photon Imaging Camera pn detector (EPIC-pn), we develop XMM-SuperRes and XMM-DeNoise - deep learning-based models that can generate enhanced SR and DN images from real observations. The models are trained on realistic XMM-Newton simulations such that XMM-SuperRes will output images with two times smaller point-spread function and with improved noise characteristics. The XMM-DeNoise model is trained to produce images with 2.5× the input exposure time from 20 to 50 ks. When tested on real images, DN improves the image quality by 8.2 per cent, as quantified by the global peak-signal-to-noise ratio. These enhanced images allow identification of features that are otherwise hard or impossible to perceive in the original or in filtered/smoothed images with traditional methods. We demonstrate the feasibility of using our deep learning models to enhance XMM-Newton X-ray images to increase their scientific value in a way that could benefit the legacy of the XMM-Newton archive.",

keywords = "techniques: high angular resolution, techniques: image processing, X-rays: general, PHOTON IMAGING CAMERA, ILLUSTRISTNG SIMULATIONS, CHANDRA, DECONVOLUTION, SIMILARITY, CLUSTERS",

author = "S.F. Sweere and I. Valtchanov and M. Lieu and A. Vojtekova and E. Verdugo and M. Santos-Lleo and F. Pacaud and A. Briassouli and {Campora Perez}, D.",

note = "Funding Information: We would like to thank the anonymous referee for the valuable and constructive comments that helped improve the paper. We also thank the XMM-Newton Project Scientist, Norbert Schartel for many helpful discussions during the initiation and the development of the study. We appreciate the valuable assistance from the SIXTE team. Based on observations obtained with XMM-Newton, an ESA science mission with instruments and contributions directly funded by ESA Member States and NASA. The IllustrisTNG simulations were undertaken with compute time awarded by the Gauss Centre for Supercomputing (GCS) under GCS Large-Scale Projects GCS- ILLU and GCS-DWAR on the GCS share of the supercomputer Hazel Hen at the High Performance Computing Center Stuttgart (HLRS), as well as on the machines of the Max Planck Computing and Data Facility (MPCDF) in Garching, Germany. SFS acknowledges support from the ESA traineeship program and further support from ESA through the Science Faculty of the European Space Astronomy Centre (ESAC) -Funding reference ESA-SCI-SC-LE- 033. ML acknowledges a Machine Learning and Cosmology research fellowship from the University of Nottingham. Publisher Copyright: {\textcopyright} 2022 The Author(s).",

year = "2022",

month = dec,

day = "1",

doi = "10.1093/mnras/stac2437",

language = "English",

volume = "517",

pages = "4054--4069",

journal = "Monthly Notices of the Royal Astronomical Society",

issn = "0035-8711",

publisher = "Oxford University Press",

number = "3",

}

TY - JOUR

T1 - Deep learning-based super-resolution and de-noising for XMM-newton images

AU - Sweere, S.F.

AU - Valtchanov, I.

AU - Lieu, M.

AU - Vojtekova, A.

AU - Verdugo, E.

AU - Santos-Lleo, M.

AU - Pacaud, F.

AU - Briassouli, A.

AU - Campora Perez, D.

N1 - Funding Information: We would like to thank the anonymous referee for the valuable and constructive comments that helped improve the paper. We also thank the XMM-Newton Project Scientist, Norbert Schartel for many helpful discussions during the initiation and the development of the study. We appreciate the valuable assistance from the SIXTE team. Based on observations obtained with XMM-Newton, an ESA science mission with instruments and contributions directly funded by ESA Member States and NASA. The IllustrisTNG simulations were undertaken with compute time awarded by the Gauss Centre for Supercomputing (GCS) under GCS Large-Scale Projects GCS- ILLU and GCS-DWAR on the GCS share of the supercomputer Hazel Hen at the High Performance Computing Center Stuttgart (HLRS), as well as on the machines of the Max Planck Computing and Data Facility (MPCDF) in Garching, Germany. SFS acknowledges support from the ESA traineeship program and further support from ESA through the Science Faculty of the European Space Astronomy Centre (ESAC) -Funding reference ESA-SCI-SC-LE- 033. ML acknowledges a Machine Learning and Cosmology research fellowship from the University of Nottingham. Publisher Copyright: © 2022 The Author(s).

PY - 2022/12/1

Y1 - 2022/12/1

N2 - The field of artificial intelligence based image enhancement has been rapidly evolving over the last few years and is able to produce impressive results on non-astronomical images. In this work, we present the first application of Machine Learning based super-resolution (SR) and de-noising (DN) to enhance X-ray images from the European Space Agency's XMM-Newton telescope. Using XMM-Newton images in band [0.5, 2] keV from the European Photon Imaging Camera pn detector (EPIC-pn), we develop XMM-SuperRes and XMM-DeNoise - deep learning-based models that can generate enhanced SR and DN images from real observations. The models are trained on realistic XMM-Newton simulations such that XMM-SuperRes will output images with two times smaller point-spread function and with improved noise characteristics. The XMM-DeNoise model is trained to produce images with 2.5× the input exposure time from 20 to 50 ks. When tested on real images, DN improves the image quality by 8.2 per cent, as quantified by the global peak-signal-to-noise ratio. These enhanced images allow identification of features that are otherwise hard or impossible to perceive in the original or in filtered/smoothed images with traditional methods. We demonstrate the feasibility of using our deep learning models to enhance XMM-Newton X-ray images to increase their scientific value in a way that could benefit the legacy of the XMM-Newton archive.

AB - The field of artificial intelligence based image enhancement has been rapidly evolving over the last few years and is able to produce impressive results on non-astronomical images. In this work, we present the first application of Machine Learning based super-resolution (SR) and de-noising (DN) to enhance X-ray images from the European Space Agency's XMM-Newton telescope. Using XMM-Newton images in band [0.5, 2] keV from the European Photon Imaging Camera pn detector (EPIC-pn), we develop XMM-SuperRes and XMM-DeNoise - deep learning-based models that can generate enhanced SR and DN images from real observations. The models are trained on realistic XMM-Newton simulations such that XMM-SuperRes will output images with two times smaller point-spread function and with improved noise characteristics. The XMM-DeNoise model is trained to produce images with 2.5× the input exposure time from 20 to 50 ks. When tested on real images, DN improves the image quality by 8.2 per cent, as quantified by the global peak-signal-to-noise ratio. These enhanced images allow identification of features that are otherwise hard or impossible to perceive in the original or in filtered/smoothed images with traditional methods. We demonstrate the feasibility of using our deep learning models to enhance XMM-Newton X-ray images to increase their scientific value in a way that could benefit the legacy of the XMM-Newton archive.

KW - techniques: high angular resolution

KW - techniques: image processing

KW - X-rays: general

KW - PHOTON IMAGING CAMERA

KW - ILLUSTRISTNG SIMULATIONS

KW - CHANDRA

KW - DECONVOLUTION

KW - SIMILARITY

KW - CLUSTERS

U2 - 10.1093/mnras/stac2437

DO - 10.1093/mnras/stac2437

M3 - Article

SN - 0035-8711

VL - 517

SP - 4054

EP - 4069

JO - Monthly Notices of the Royal Astronomical Society

JF - Monthly Notices of the Royal Astronomical Society

IS - 3

ER -