A method for asteroid detection using convolutional neural networks on VST images

¹ Laboratory of Astrophysics, École Polytechnique Fédérale de Lausanne, Chemin Pegasi 51, 1290 Versoix, Switzerland
² European Space Agency/ESAC, Camino Bajo del Castillo s/n, 28692 Madrid, Spain
³ Fysikum, Stockholm University, 106 91 Stockholm, Sweden
⁴ Université de Strasbourg, CNRS, Observatoire astronomique de Strasbourg, UMR 7550, 67000 Strasbourg, France
⁵ European Southern Observatory, Karl-Schwarzschild-Strasse 2, 85748 Garching bei München, Germany
⁶ European Space Agency/ESRIN PDO NEO Coordination Centre, Largo Galileo Galilei 1, 00044 Frascati, Roma, Italy
⁷ European Space Agency/ESAC PDO NEO Coordination Centre, Camino Bajo del Castillo s/n, 28692 Madrid, Spain
⁸ Institut de Ciències del Cosmos, Universitat de Barcelona, Martí i Franquès 1, 08028 Barcelona, Spain
⁹ ICREA, Pg. Lluís Companys 23, Barcelona, 08010 Spain
¹⁰ European Space Agency/ESRIN, Largo Galileo Galilei 1, 00044 Frascati, Roma, Italy
¹¹ Institute for Particle Physics and Astrophysics, ETH Zurich, Wolfgang-Pauli-Strasse 27, 8093 Zurich, Switzerland
¹² Kapteyn Astronomical Institute, University of Groningen, 9700 AV Groningen, The Netherlands
¹³ Swiss Data Science Center and CVLab, École Polytechnique Fédérale de Lausanne, 1015 Lausanne, Switzerland

^★ Corresponding author; belen.irureta@epfl.ch

Received: 25 October 2024
Accepted: 9 January 2025

Abstract

Context. The study of asteroids, particularly near-Earth asteroids, is key to gaining insights into our Solar System and can help prevent dangerous collisions. Beyond finding new objects, additional observations of known asteroids will improve our knowledge of their orbit.

Aims. We have developed an automated pipeline to process and search for asteroid trails in images taken with OmegaCAM, the wide- field imager mounted on the VLT Survey Telescope (VST), on the European Southern Observatory’s Cerro Paranal. The pipeline inputs a FITS image and outputs the position, length, and angle of all the asteroids trails detected.

Methods. A convolutional neural network was trained on a set of synthetic asteroid trails, with trail lengths 5–120 pixels (1–25″) and S/Ns 3–20. Its performance was tested on synthetic trails and validated using real trails, chosen from the Solar System Object Image Search of the Canadian Astronomy Data Centre.

Results. On the synthetic trails, the pipeline achieved a completeness of 70% for trails with length ≥15 pixels (3″), with a precision of 82%. On the real trails, the pipeline achieved a completeness of 65%, with a precision of 44%, a lower value likely due to the higher presence of contaminants and stars in the field. The pipeline was able to detect both low- and high-S/N asteroid trails.

Conclusions. Our method shows a strong potential to make new discoveries and precoveries in VST data across the S/N range studied, especially in the fainter end, which remains largely unexplored.