Abstract:Filaments host star formation and are fundamental structures of galaxies. Their diversity, as observed in the interstellar medium, from very low-density structures to very dense hubs, and their complex life cycles make their complete detection challenging over this large diversity range. Using 2D H$_2$ column density images obtained as part of the Herschel Hi-GAL survey of the Galactic plane (Gp), we want to detect, simultaneously and using a single model, filaments over a large range of column density and contrast over the whole Gp. In particular, we target low-contrast and low-density structures that are particularly difficult to detect with classical algorithms. The whole H$_2$ column density image of the Gp was subdivided into individual patches of 32times 32 pixels. Following our proof of concept study aimed at exploring the potential of supervised learning for the detection of filaments, we propose an innovative supervised learning method based on adding information by encoding the position of these patches in the Gp. To allow the segmentation of the whole Gp, we introduced a random procedure that preserves the balance within the model training and testing datasets over the Gp plane. Four architectures and six models were tested and compared using different metrics. For the first time, a segmentation of the whole Gp has been obtained using supervised deep learning. A comparison of the models based on metrics and astrophysical results shows that one of the architectures (PE-UNet-Latent), where the position encoding was done in the latent space gives the best performance to detect filaments over the whole range of density and contrast observed in the Gp. A normalized map of the whole Gp was also produced and reveals the highly filamentary structure of the Gp in all density regimes. We successfully tested the generalization of our best model by applying it to the 2D 12CO COHRS molecular data obtained on a 52 portion (in longitude) of the plane. We demonstrate the interest of position encoding to allow the detection of filaments over the wide range of density and contrast observed in the Gp. The produced maps (both normalized and segmented) offer a unique opportunity for follow-up studies of the life cycle of Galactic filaments. The promising generalization possibility tested on a molecular dataset of the Gp opens new opportunities for systematic detection of filamentary structures in the big data context available for the Gp.

A universal method for solar filament detection from H-alpha observations using semi-supervised deep learning

A Universal Method for Solar Filament Detection from H-alpha Observations using Semi-supervised Deep Learning

Solar Filament Recognition Based on Deep Learning

Solar-Filament Detection and Classification Based on Deep Learning

Developing an Advanced Automated Method for Solar Filament Recognition and Its Scientific Application to a Solar Cycle of MLSO Hα Data

Development of an Advanced Automated Method for Solar Filament Recognition and Its Scientific Application to a Solar Cycle of MLSO Hα Data

Automated High-Precision Recognition of Solar Filaments Based on an Improved U2-Net

Developing an Automated Detection, Tracking, and Analysis Method for Solar Filaments Observed by CHASE via Machine Learning

Developing an Automated Detection, Tracking and Analysis Method for Solar Filaments Observed by CHASE via Machine Learning

Automated Detection of Filaments and their Disappearance using Full Disc H-alpha Images

Automated Detection Methods for Solar Activities and an Application for Statistic Analysis of Solar Filament

Detection of Solar Filaments Using Suncharts from Kodaikanal Solar Observatory Archive Employing a Clustering Approach

Solar Filament Segmentation Based on Improved U-Nets

Tracing Halpha Fibrils through Bayesian Deep Learning

A dataset of manually annotated filaments from H-alpha observations

Long term study of the solar filaments from the Synoptic Maps as derived from $H_α$ Spectroheliograms of Kodaikanal Observatory

Ultra Low-Frequency Oscillations of a Solar Filament in Hα Revealed With the Data of the Global Oscillation Network Group (Gong)

Statistical Analysis of Filament Features Based on the Hα Solar Images from 1988 to 2013 by Computer Automated Detection Method

Improvements on coronal hole detection in SDO/AIA images using supervised classification

A New Comprehensive Data Set of Solar Filaments of 100 yr Interval. I.

Supervised machine learning on Galactic filaments. II. Encoding the position to optimize the detection of filaments over a wide range of column density and contrast