A Convolutional Neural Network–Based Approach for Detecting Solar System Objects in Wide-field Imaging

We present a deep learning method that utilizes convolutional neural networks (CNNs) to discover trans-Neptunian objects (TNOs) in wide-field survey imaging data. Our CNNs were trained using artificial sources planted into a time series of 44 205 s CFHT MegaCam large-format mosaic images. We extract...

Full description

Saved in:
Bibliographic Details
Main Authors: Aram Lee, J. J. Kavelaars, Hossen Teimoorinia, Wesley Fraser, Edward Ashton
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:The Planetary Science Journal
Subjects:
Online Access:https://doi.org/10.3847/PSJ/add409
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849425058669264896
author Aram Lee
J. J. Kavelaars
Hossen Teimoorinia
Wesley Fraser
Edward Ashton
author_facet Aram Lee
J. J. Kavelaars
Hossen Teimoorinia
Wesley Fraser
Edward Ashton
author_sort Aram Lee
collection DOAJ
description We present a deep learning method that utilizes convolutional neural networks (CNNs) to discover trans-Neptunian objects (TNOs) in wide-field survey imaging data. Our CNNs were trained using artificial sources planted into a time series of 44 205 s CFHT MegaCam large-format mosaic images. We extracted 64 × 64 pixel subimages and labeled each subimage pair with the presence or absence of a moving source and the source’s location and magnitude. Our MobileNet-derived classification model achieved 91% recall with 90% precision on test data. A separate regression model predicted the source locations with a mean absolute error of ±1.5 pixels for sources with a signal-to-noise ratio (SNR) of 16.5 or higher. By grouping sources based on linear sky motion, we achieved ∼40% detection limit for sources with an SNR ≈ 7 in individual frames. We also examined a scoring approach to reduce the false positives. In this approach, images are scored based on the probability values from the classification model. With the scoring approach, we achieved a detection limit of SNR ≈ 3 in individual frames. We detected approximately 200 solar system object (SSO) candidates ( sim 5 moving rates consistent with TNOs) in our 1 square degree of imaging. We tested the trained models on images from different sky regions, confirming that the models learned from the motion of sources rather than from the backgrounds or shapes of sources. We demonstrate that deep learning object detection algorithms can aid in TNO and SSO discovery, supporting future discovery pipeline development.
format Article
id doaj-art-bcb8a4ccc64740b28644a6945db30f0b
institution Kabale University
issn 2632-3338
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series The Planetary Science Journal
spelling doaj-art-bcb8a4ccc64740b28644a6945db30f0b2025-08-20T03:29:53ZengIOP PublishingThe Planetary Science Journal2632-33382025-01-016614710.3847/PSJ/add409A Convolutional Neural Network–Based Approach for Detecting Solar System Objects in Wide-field ImagingAram Lee0https://orcid.org/0000-0002-4258-2964J. J. Kavelaars1https://orcid.org/0000-0001-7032-5255Hossen Teimoorinia2https://orcid.org/0000-0002-4348-6974Wesley Fraser3https://orcid.org/0000-0001-6680-6558Edward Ashton4https://orcid.org/0000-0002-4637-8426University of Victoria , Department of Physics and Astronomy, Elliott Building, 3800 Finnerty Road, Victoria, BC, V8P 5C2, Canada; Department of Physics and Astronomy, University of British Columbia , 6224 Agricultural Road, Vancouver, BC V6T 1Z1, CanadaUniversity of Victoria , Department of Physics and Astronomy, Elliott Building, 3800 Finnerty Road, Victoria, BC, V8P 5C2, Canada; Department of Physics and Astronomy, University of British Columbia , 6224 Agricultural Road, Vancouver, BC V6T 1Z1, Canada; National Research Council of Canada , Herzberg Astronomy and Astrophysics Research Centre, 5071 West Saanich Road, Victoria, BC, V9E 2E7, CanadaUniversity of Victoria , Department of Physics and Astronomy, Elliott Building, 3800 Finnerty Road, Victoria, BC, V8P 5C2, Canada; National Research Council of Canada , Herzberg Astronomy and Astrophysics Research Centre, 5071 West Saanich Road, Victoria, BC, V9E 2E7, CanadaUniversity of Victoria , Department of Physics and Astronomy, Elliott Building, 3800 Finnerty Road, Victoria, BC, V8P 5C2, Canada; National Research Council of Canada , Herzberg Astronomy and Astrophysics Research Centre, 5071 West Saanich Road, Victoria, BC, V9E 2E7, CanadaInstitute of Astronomy and Astrophysics , Academia Sinica, No.1, Sec. 4, Roosevelt Road, Taipei, 10617, TaiwanWe present a deep learning method that utilizes convolutional neural networks (CNNs) to discover trans-Neptunian objects (TNOs) in wide-field survey imaging data. Our CNNs were trained using artificial sources planted into a time series of 44 205 s CFHT MegaCam large-format mosaic images. We extracted 64 × 64 pixel subimages and labeled each subimage pair with the presence or absence of a moving source and the source’s location and magnitude. Our MobileNet-derived classification model achieved 91% recall with 90% precision on test data. A separate regression model predicted the source locations with a mean absolute error of ±1.5 pixels for sources with a signal-to-noise ratio (SNR) of 16.5 or higher. By grouping sources based on linear sky motion, we achieved ∼40% detection limit for sources with an SNR ≈ 7 in individual frames. We also examined a scoring approach to reduce the false positives. In this approach, images are scored based on the probability values from the classification model. With the scoring approach, we achieved a detection limit of SNR ≈ 3 in individual frames. We detected approximately 200 solar system object (SSO) candidates ( sim 5 moving rates consistent with TNOs) in our 1 square degree of imaging. We tested the trained models on images from different sky regions, confirming that the models learned from the motion of sources rather than from the backgrounds or shapes of sources. We demonstrate that deep learning object detection algorithms can aid in TNO and SSO discovery, supporting future discovery pipeline development.https://doi.org/10.3847/PSJ/add409Solar systemKuiper beltDetectionNeural networks
spellingShingle Aram Lee
J. J. Kavelaars
Hossen Teimoorinia
Wesley Fraser
Edward Ashton
A Convolutional Neural Network–Based Approach for Detecting Solar System Objects in Wide-field Imaging
The Planetary Science Journal
Solar system
Kuiper belt
Detection
Neural networks
title A Convolutional Neural Network–Based Approach for Detecting Solar System Objects in Wide-field Imaging
title_full A Convolutional Neural Network–Based Approach for Detecting Solar System Objects in Wide-field Imaging
title_fullStr A Convolutional Neural Network–Based Approach for Detecting Solar System Objects in Wide-field Imaging
title_full_unstemmed A Convolutional Neural Network–Based Approach for Detecting Solar System Objects in Wide-field Imaging
title_short A Convolutional Neural Network–Based Approach for Detecting Solar System Objects in Wide-field Imaging
title_sort convolutional neural network based approach for detecting solar system objects in wide field imaging
topic Solar system
Kuiper belt
Detection
Neural networks
url https://doi.org/10.3847/PSJ/add409
work_keys_str_mv AT aramlee aconvolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT jjkavelaars aconvolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT hossenteimoorinia aconvolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT wesleyfraser aconvolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT edwardashton aconvolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT aramlee convolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT jjkavelaars convolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT hossenteimoorinia convolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT wesleyfraser convolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging
AT edwardashton convolutionalneuralnetworkbasedapproachfordetectingsolarsystemobjectsinwidefieldimaging