3D detection and characterization of ALMA sources through deep learning

Abstract

We present a Deep-Learning (DL) pipeline developed for the detection and characterization of astronomical sources within simulated Atacama Large Millimeter/submillimeter Array (ALMA) data cubes. The pipeline is composed of six DL models: a Convolutional Autoencoder for source detection within the spatial domain of the integrated data cubes, a Recurrent Neural Network (RNN) for denoising and peak detection within the frequency domain, and four Residual Neural Networks (ResNets) for source characterization. The combination of spatial and frequency information improves completeness while decreasing spurious signal detection. To train and test the pipeline, we developed a simulation algorithm able to generate realistic ALMA observations, i.e. both sky model and dirty cubes. The algorithm simulates always a central source surrounded by fainter ones scattered within the cube. Some sources were spatially superimposed in order to test the pipeline deblending capabilities. The detection performances of the pipeline were compared to those of other methods and significant improvements in performances were achieved. Source morphologies are detected with subpixel accuracies obtaining mean residual errors of 10−3 pixel (0.1 mas) and 10−1 mJy/beam on positions and flux estimations, respectively. Projection angles and flux densities are also recovered within $10\%$ of the true values for $80\%$ and $73\%$ of all sources in the test set, respectively. While our pipeline is fine-tuned for ALMA data, the technique is applicable to other interferometric observatories, as SKA, LOFAR, VLBI, and VLTI.

A BRAIN Study to Tackle Image Analysis with Artificial Intelligence in the ALMA 2030 Era

“…In this study, the information field theory [8] approach RESOLVE [9][10][11] and the astroinformatics technique DeepFocus [12] are applied to ALMA-calibrated data. A companion package of DeepFocus, named ALMASim, was developed with the additional scope to provide simulated data as well as open software for the scientific community.…”

Regularized Maximum Likelihood Image Synthesis and Validation for ALMA Continuum Observations of Protoplanetary Disks

“…Many of the drawbacks encountered with CLEAN can be partially or completely avoided by using an alternative class of imaging techniques which incorporate additional information into the image synthesis routine through the use of regularizers. These methods have been successful across a diverse array of methodologies and implementations, including maximum entropy methods (MEM, e.g., Ponsonby 1973;Ables 1974;Cornwell & Evans 1985;Narayan & Nityananda 1986;Casassus et al 2013), compressed sensing and sparse reconstruction methods (e.g., Wiaux et al 2009;Li et al 2011;Dabbech et al 2015;Onose et al 2016), visibility model fitting (e.g., Tazzari et al 2018;Jennings et al 2020), or machine learning-based methods (e.g., Dabbech et al 2022;Sanchez-Bermudez et al 2022;Delli Veneri et al 2023;Terris et al 2023).…”

Bayesian and Machine Learning Methods in the Big Data Era for Astronomical Imaging

“…The DL pipeline [20] (paper submitted) is developed to detect and characterize sources in ALMA cubes. The idea is to learn the underlying features from the ALMA dirty image cubes (I D ).…”