Leveraging Movement Representation from Contrastive Learning for Asteroid Detection

Simple item page
dc.contributor.authorKongsathitporn N.
dc.contributor.authorSupratak A.
dc.contributor.authorNoysena K.
dc.contributor.authorAwiphan S.
dc.contributor.authorSteeghs D.
dc.contributor.authorPollacco D.
dc.contributor.authorUlaczyk K.
dc.contributor.authorLyman J.
dc.contributor.authorAckley K.
dc.contributor.authorO’Neill D.
dc.contributor.authorKumar A.
dc.contributor.authorGalloway D.K.
dc.contributor.authorJiménez-Ibarra F.
dc.contributor.authorDhillon V.S.
dc.contributor.authorDyer M.J.
dc.contributor.authorO’Brien P.
dc.contributor.authorRamsay G.
dc.contributor.authorPallé E.
dc.contributor.authorKotak R.
dc.contributor.authorKillestein T.L.
dc.contributor.authorNuttall L.K.
dc.contributor.authorBreton R.P.
dc.contributor.correspondenceKongsathitporn N.
dc.contributor.otherMahidol University
dc.date.accessioned2025-01-10T18:04:29Z
dc.date.available2025-01-10T18:04:29Z
dc.date.issued2024-12-01
dc.description.abstractTo 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.
dc.identifier.citationPublications of the Astronomical Society of the Pacific Vol.136 No.12 (2024)
dc.identifier.doi10.1088/1538-3873/ad8c83
dc.identifier.issn00046280
dc.identifier.scopus2-s2.0-85213845589
dc.identifier.urihttps://repository.li.mahidol.ac.th/handle/20.500.14594/102660
dc.rights.holderSCOPUS
dc.subjectEarth and Planetary Sciences
dc.subjectPhysics and Astronomy
dc.titleLeveraging Movement Representation from Contrastive Learning for Asteroid Detection
dc.typeArticle
mu.datasource.scopushttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85213845589&origin=inward
oaire.citation.issue12
oaire.citation.titlePublications of the Astronomical Society of the Pacific
oaire.citation.volume136
oairecerif.author.affiliationNational Astronomical Research Institute of Thailand
oairecerif.author.affiliationFaculty of Science, Engineering and Medicine
oairecerif.author.affiliationUniversity of Leicester
oairecerif.author.affiliationUniversity of Portsmouth
oairecerif.author.affiliationMonash University
oairecerif.author.affiliationArmagh Observatory
oairecerif.author.affiliationMahidol University
oairecerif.author.affiliationInstituto Astrofisico de Canarias
oairecerif.author.affiliationTurun yliopisto
oairecerif.author.affiliationThe University of Manchester
oairecerif.author.affiliationThe University of Sheffield
Files
Collections
Scopus 2024