Supervised machine learning on Galactic filaments Revealing the filamentary structure of the Galactic interstellar medium

Abstract

Context. Filaments are ubiquitous in the Galaxy, and they host star formation. Detecting them in a reliable way is therefore key towards our understanding of the star formation process. Aims. We explore whether supervised machine learning can identify filamentary structures on the whole Galactic plane. Methods. We used two versions of UNet-based networks for image segmentation.We used H2 column density images of the Galactic plane obtained with Herschel Hi-GAL data as input data. We trained the UNet-based networks with skeletons (spine plus branches) of filaments that were extracted from these images, together with background and missing data masks that we produced. We tested eight training scenarios to determine the best scenario for our astrophysical purpose of classifying pixels as filaments. Results. The training of the UNets allows us to create a new image of the Galactic plane by segmentation in which pixels belonging to filamentary structures are identified. With this new method, we classify more pixels (more by a factor of 2 to 7, depending on the classification threshold used) as belonging to filaments than the spine plus branches structures we used as input. New structures are revealed, which are mainly low-contrast filaments that were not detected before.We use standard metrics to evaluate the performances of the different training scenarios. This allows us to demonstrate the robustness of the method and to determine an optimal threshold value that maximizes the recovery of the input labelled pixel classification. Conclusions. This proof-of-concept study shows that supervised machine learning can reveal filamentary structures that are present throughout the Galactic plane. The detection of these structures, including low-density and low-contrast structures that have never been seen before, offers important perspectives for the study of these filaments.

27 pages, 22 figures, accepted by Astronomy & Astrophysics

References in corpus (14)

• The identification of filaments on far infrared and submillimiter images. Morphology, physical conditions and relation with star formation of filamentary structure
• Properties of the dense core population in Orion B as seen by the Herschel Gould Belt survey
• HCN/HNC intensity ratio: a new chemical thermometer for the molecular ISM
• The Hi-GAL catalogue of dusty filamentary structures in the Galactic Plane
• Multiscale, multiwavelength extraction of sources and filaments using separation of the structural components: getsf
• Probing fragmentation and velocity sub-structure in the massive NGC 6334 filament with ALMA
• Deep learning approach for identification of HII regions during reionization in 21-cm observations
• The role of Galactic HII regions in the formation of filaments. High-resolution submilimeter imaging of RCW 120 with ArTéMiS
• The hierarchical fragmentation of filaments and the role of sub-filaments
• Rotating filament in Orion B: Do cores inherit their angular momentum from their parent filament?
• Merging Filaments I: A race against collapse
• The widths of magnetised filaments in molecular clouds
• Quantifying the fine structures of disk galaxies with deep learning:Segmentation of spiral arms in different Hubble types
• Citizen ASAS-SN Data Release I: Variable Star Classification Using Citizen Science

Cited by in corpus (6)

• The problem of dust attenuation in photometric decomposition of edge-on galaxies and possible solutions
• A Deep Neural Network Approach to Compact Source Removal
• Comparing the morphology of molecular clouds without supervision
• Analysis of Galactic cirrus filaments in HSC-SSP high-resolution deep images using artificial neural networks
• Sutra : An integrated framework for identification and characterization of filaments in the interstellar medium
• Interstellar filament detection and characterization: methods and implications for studies of magnetized interstellar medium