A Deep Neural Network Approach to Compact Source Removal

Abstract

Analysing extended emission in photometric observations of star-forming regions requires maps free from compact foreground, embedded, and background sources, which can interfere with various techniques used to characterise the interstellar medium. Within the framework of the NEMESIS project, we apply machine learning techniques to improve our understanding of the star formation timescales, which involves the unbiased analysis of the extended emission in these regions. We present a deep learning-based method for separating the signals of compact sources and extended emission in photometric observations made by the Herschel Space Observatory, facilitating the analysis of extended emission and improving the photometry of compact sources. Central to our approach is a modified U-Net architecture with partial convolutional layers. This method enables effective source removal and background estimation across various flux densities, using a series of partial convolutional layers, batch normalisation, and ReLU activation layers within blocks. Our training process utilised simulated sources injected into Herschel images, with controlled flux densities against known backgrounds. A dynamic, signal-to-noise ratio-based adaptive masking system was implemented to assess how prominently a compact source stands out from the surrounding background. The results demonstrate that our method can significantly improve the photometric accuracy in the presence of highly fluctuating backgrounds. Moreover, the approach can preserve all characteristics of the images, including the noise properties. The presented approach allows users to analyse extended emission without the interference of disturbing point sources or perform more precise photometry of sources located in complex environments. We also provide a Python tool with tutorials and examples to help the community effectively utilise this method.

23 pages, 13 figures, 10 tables, submitted to Astronomy&Astrophysics

References in corpus (24)

• Array Programming with NumPy
• Astropy: A Community Python Package for Astronomy
• The Astropy Project: Sustaining and Growing a Community-oriented Open-source Project and the Latest Major Release (v5.0) of the Core Package
• Binary companions of evolved stars in APOGEE DR14: Search method and catalog of ~5,000 companions
• Herschel Space Observatory - An ESA facility for far-infrared and submillimetre astronomy
• Theory of Star Formation
• The Photodetector Array Camera and Spectrometer (PACS) on the Herschel Space Observatory
• Control of star formation by supersonic turbulence
• The Star Formation Rate of Turbulent Magnetized Clouds: Comparing Theory, Simulations, and Observations
• The Star Formation Rate and Gas Surface Density Relation in the Milky Way: Implications for Extragalactic Studies
• Cores, filaments, and bundles: hierarchical core formation in the L1495/B213 Taurus region
• The Physics of Star Formation
• An ALMA study of the Orion Integral Filament: I. Evidence for narrow fibers in a massive cloud
• The Herschel-PACS photometer calibration: Point-source flux calibration for scan maps
• Source extraction and photometry for the far-infrared and sub-millimeter continuum in the presence of complex backgrounds
• Multiscale, multiwavelength extraction of sources and filaments using separation of the structural components: getsf
• CARMA Large Area Star Formation Survey: Structure and Kinematics of Dense Gas in Serpens Main
• A possible link between the power spectrum of interstellar filaments and the origin of the prestellar core mass function
• Characterizing the structure of diffuse emission in Hi-GAL maps
• Supervised machine learning on Galactic filaments Revealing the filamentary structure of the Galactic interstellar medium
• The new Herschel/PACS Point Source Catalogue
• Emergence of high-mass stars in complex fiber networks (EMERGE). I. Early ALMA Survey: observations and massive data reduction
• Herschel-PACS photometry of faint stars
• First results with the boloSource() algorithm: Photometry of faint standard stars observed by Herschel/PACS

Cited by in corpus (2)

• The NEMESIS Catalogue of Young Stellar Objects for the Orion Star Formation Complex. I. General description of data curation
• Latent-space Field Tension for Astrophysical Component Detection An application to X-ray imaging