A1 Refereed original research article in a scientific journal
Leveraging Movement Representation from Contrastive Learning for Asteroid Detection
Authors: Kongsathitporn, Noppachanin; Supratak, Akara; Noysena, Kanthanakorn; Awiphan, Supachai; Steeghs, Danny; Pollacco, Don; Ulaczyk, Krzysztof; Lyman, Joseph; Ackley, Kendall; O'Neill, David; Kumar, Amit; Galloway, Duncan K.; Jiménez-Ibarra, Felipe; Dhillon, Vik. S.; Dyer, Martin J.; O'Brien, Paul; Ramsay, Gavin; Palle, Enric; Kotak, Rubin; Killestein, Thomas L.; Nuttall, Laura K.; Breton, Rene P.
Publisher: IOP Publishing Ltd
Publishing place: BRISTOL
Publication year: 2024
Journal: Publications of the Astronomical Society of the Pacific
Journal name in source: PUBLICATIONS OF THE ASTRONOMICAL SOCIETY OF THE PACIFIC
Journal acronym: PUBL ASTRON SOC PAC
Article number: 124507
Volume: 136
Issue: 12
Number of pages: 20
ISSN: 0004-6280
eISSN: 1538-3873
DOI: https://doi.org/10.1088/1538-3873/ad8c83
Web address : https://iopscience.iop.org/article/10.1088/1538-3873/ad8c83
Self-archived copy’s web address: https://research.utu.fi/converis/portal/detail/Publication/478071830
To support asteroid-related studies, current motion detectors are utilized to select moving object candidates based on their visualizations and movements in sequences of sky exposures. However, the existing detectors encounter the manual parameter settings which require experts to assign proper parameters. Moreover, although the deep learning approach could automate the detection process, these approaches still require synthetic images and hand-engineered features to improve their performance. In this work, we propose an end-to-end deep learning model consisting of two branches. The first branch is trained with contrastive learning to extract a contrastive feature from sequences of sky exposures. This learning method encourages the model to capture a lower-dimensional representation, ensuring that sequences with moving sources (i.e., potential asteroids) are distinct from those without moving sources. The second branch is designed to learn additional features from the sky exposure sequences, which are then concatenated into the movement features before being processed by subsequent layers for the detection of asteroid candidates. We evaluate our model on sufficiently long-duration sequences and perform a comparative study with detection software. Additionally, we demonstrate the use of our model to suggest potential asteroids using photometry filtering. The proposed model outperforms the baseline model for asteroid streak detection by +7.70% of f1-score. Moreover, our study shows promising performance for long-duration sequences and improvement after adding the contrastive feature. Additionally, we demonstrate the uses of our model with the filtering to detect potential asteroids in wide-field detection using the long-duration sequences. Our model could complement the software as it suggests additional asteroids to its detection result.
Downloadable publication This is an electronic reprint of the original article. |  |
Funding information in the publication:
This project is funded by National Research Council of Thailand (NRCT). This work was also supported by a National Astronomical Research Institute of Thailand (NARIT) and Thailand Science Research and Innovation (TSRI) research grant. We are also grateful to Assoc. Prof. Dr. Suppawong Tuarob and Asst. Prof. Dr. Thanapon Noraset from the Faculty of Information and Communication Technology, Mahidol University for their suggestions, and moral support. The Gravitational-wave Optical Transient Observer (GOTO) project acknowledges the support of the Monash-Warwick Alliance; University of Warwick; Monash University; University of Sheffield; University of Leicester; Armagh Observatory & Plane- tarium; the National Astronomical Research Institute of Thailand (NARIT); Instituto de Astrof´ısica de Canarias (IAC); University of Portsmouth; University of Turku. We acknowledge financial support from the Agencia Estatal de Investigaci\'on of the Ministerio de Ciencia e Innovaci\'on MCIN/AEI/10.13039/501100011033 and the ERDF "A way of making Europe" through project PID2021-125627OB-C32, and from the Centre of Excellence "Severo Ochoa" award to the Instituto de Astrofisica de Canarias.
