{"title":"Vision Transformers for identifying asteroids interacting with secular resonances","authors":"","doi":"10.1016/j.icarus.2024.116346","DOIUrl":null,"url":null,"abstract":"<div><div>Currently, more than 1.4 million asteroids are known in the main belt. Future surveys, like those that the Vera C. Rubin Observatory will perform, may increase this number to up to 8 million. While in the past identification of asteroids interacting with secular resonances was performed by a visual analysis of images of resonant arguments, this method is no longer feasible in the age of big data. Deep learning methods based on Convolutional Neural Networks (CNNs) have been used in the recent past to automatically classify databases of several thousands of images of resonant arguments for resonances like the <span><math><msub><mrow><mi>ν</mi></mrow><mrow><mn>6</mn></mrow></msub></math></span>, the <span><math><mrow><mi>g</mi><mo>−</mo><mn>2</mn><msub><mrow><mi>g</mi></mrow><mrow><mn>6</mn></mrow></msub><mo>+</mo><msub><mrow><mi>g</mi></mrow><mrow><mn>5</mn></mrow></msub></mrow></math></span>, and the <span><math><mrow><mi>s</mi><mo>−</mo><msub><mrow><mi>s</mi></mrow><mrow><mn>6</mn></mrow></msub><mo>−</mo><msub><mrow><mi>g</mi></mrow><mrow><mn>5</mn></mrow></msub><mo>+</mo><msub><mrow><mi>g</mi></mrow><mrow><mn>6</mn></mrow></msub></mrow></math></span>. However, it has been shown that computer vision methods based on the Transformer architecture tend to outperform CNN models if the scale of the image database is large enough. Here, for the first time, we developed a Vision Transformer (ViT) model and applied it to publicly available databases for the three secular resonances quoted above. ViT architecture outperforms CNN models in speed and accuracy while avoiding overfitting concerns. If hyper-parameter tuning research is undertaken for each analyzed database, ViT models should be preferred over CNN architectures.</div></div>","PeriodicalId":13199,"journal":{"name":"Icarus","volume":null,"pages":null},"PeriodicalIF":2.5000,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Icarus","FirstCategoryId":"101","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0019103524004068","RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ASTRONOMY & ASTROPHYSICS","Score":null,"Total":0}
引用次数: 0
Abstract
Currently, more than 1.4 million asteroids are known in the main belt. Future surveys, like those that the Vera C. Rubin Observatory will perform, may increase this number to up to 8 million. While in the past identification of asteroids interacting with secular resonances was performed by a visual analysis of images of resonant arguments, this method is no longer feasible in the age of big data. Deep learning methods based on Convolutional Neural Networks (CNNs) have been used in the recent past to automatically classify databases of several thousands of images of resonant arguments for resonances like the , the , and the . However, it has been shown that computer vision methods based on the Transformer architecture tend to outperform CNN models if the scale of the image database is large enough. Here, for the first time, we developed a Vision Transformer (ViT) model and applied it to publicly available databases for the three secular resonances quoted above. ViT architecture outperforms CNN models in speed and accuracy while avoiding overfitting concerns. If hyper-parameter tuning research is undertaken for each analyzed database, ViT models should be preferred over CNN architectures.
期刊介绍:
Icarus is devoted to the publication of original contributions in the field of Solar System studies. Manuscripts reporting the results of new research - observational, experimental, or theoretical - concerning the astronomy, geology, meteorology, physics, chemistry, biology, and other scientific aspects of our Solar System or extrasolar systems are welcome. The journal generally does not publish papers devoted exclusively to the Sun, the Earth, celestial mechanics, meteoritics, or astrophysics. Icarus does not publish papers that provide "improved" versions of Bode''s law, or other numerical relations, without a sound physical basis. Icarus does not publish meeting announcements or general notices. Reviews, historical papers, and manuscripts describing spacecraft instrumentation may be considered, but only with prior approval of the editor. An entire issue of the journal is occasionally devoted to a single subject, usually arising from a conference on the same topic. The language of publication is English. American or British usage is accepted, but not a mixture of these.