AstroYOLO:一种针对蓝色水平分支恒星的CNN-Transformer混合深度学习目标检测模型

IF 2.2 4区 物理与天体物理 Q2 ASTRONOMY & ASTROPHYSICS
Yuchen He, Jingjing Wu, Wenyu Wang, Bin Jiang, Yanxia Zhang
{"title":"AstroYOLO:一种针对蓝色水平分支恒星的CNN-Transformer混合深度学习目标检测模型","authors":"Yuchen He, Jingjing Wu, Wenyu Wang, Bin Jiang, Yanxia Zhang","doi":"10.1093/pasj/psad071","DOIUrl":null,"url":null,"abstract":"Abstract Blue horizontal-branch stars (BHBs) are ideal tracers for studying the Milky Way (MW) due to their bright and nearly constant magnitude. However, an incomplete screen of BHBs from a survey would result in bias of estimation of the structure or mass of the MW. With surveys of large sky telescopes like the Sloan Digital Sky Survey (SDSS), it is possible to obtain a complete sample. Thus, detecting BHBs from massive photometric images quickly and effectually is necessary. The current acquisition methods of BHBs are mainly based on manual or semi-automatic modes. Therefore, novel approaches are required to replace manual or traditional machine-learning detection. The mainstream deep-learning-based object-detection methods are often vanilla convolutional neural networks whose ability to extract global features is limited by the receptive field of the convolution operator. Recently, a new Transformer-based method has benefited from the global receptive field advantage brought by the self-attention mechanism, exceeded the vanilla convolution model in many tasks, and achieved excellent results. Therefore, this paper proposes a hybrid convolution and Transformer model called AstroYOLO to take advantage of the convolution in local feature representation and Transformer’s easier discovery of long-distance feature dependences. We conduct a comparative experiment on the 4799 SDSS DR16 photometric image dataset. The experimental results show that our model achieves 99.25% AP@50, 93.79% AP@75, and 64.45% AP@95 on the test dataset, outperforming the YOLOv3 and YOLOv4 object-detection models. In addition, we test on larger cutout images based on the same resolution. Our model can reach 99.02% AP@50, 92.00% AP@75, and 61.96% AP@95 respectively, still better than YOLOv3 and YOLOv4. These results also suggest that an appropriate size for cutout images is necessary for the performance and computation of object detection. Compared with the previous models, our model has achieved satisfactory object-detection results and can effectively improve the accuracy of BHB detection.","PeriodicalId":20733,"journal":{"name":"Publications of the Astronomical Society of Japan","volume":"48 6","pages":"0"},"PeriodicalIF":2.2000,"publicationDate":"2023-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"AstroYOLO: A hybrid CNN–Transformer deep-learning object-detection model for blue horizontal-branch stars\",\"authors\":\"Yuchen He, Jingjing Wu, Wenyu Wang, Bin Jiang, Yanxia Zhang\",\"doi\":\"10.1093/pasj/psad071\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract Blue horizontal-branch stars (BHBs) are ideal tracers for studying the Milky Way (MW) due to their bright and nearly constant magnitude. However, an incomplete screen of BHBs from a survey would result in bias of estimation of the structure or mass of the MW. With surveys of large sky telescopes like the Sloan Digital Sky Survey (SDSS), it is possible to obtain a complete sample. Thus, detecting BHBs from massive photometric images quickly and effectually is necessary. The current acquisition methods of BHBs are mainly based on manual or semi-automatic modes. Therefore, novel approaches are required to replace manual or traditional machine-learning detection. The mainstream deep-learning-based object-detection methods are often vanilla convolutional neural networks whose ability to extract global features is limited by the receptive field of the convolution operator. Recently, a new Transformer-based method has benefited from the global receptive field advantage brought by the self-attention mechanism, exceeded the vanilla convolution model in many tasks, and achieved excellent results. Therefore, this paper proposes a hybrid convolution and Transformer model called AstroYOLO to take advantage of the convolution in local feature representation and Transformer’s easier discovery of long-distance feature dependences. We conduct a comparative experiment on the 4799 SDSS DR16 photometric image dataset. The experimental results show that our model achieves 99.25% AP@50, 93.79% AP@75, and 64.45% AP@95 on the test dataset, outperforming the YOLOv3 and YOLOv4 object-detection models. In addition, we test on larger cutout images based on the same resolution. Our model can reach 99.02% AP@50, 92.00% AP@75, and 61.96% AP@95 respectively, still better than YOLOv3 and YOLOv4. These results also suggest that an appropriate size for cutout images is necessary for the performance and computation of object detection. Compared with the previous models, our model has achieved satisfactory object-detection results and can effectively improve the accuracy of BHB detection.\",\"PeriodicalId\":20733,\"journal\":{\"name\":\"Publications of the Astronomical Society of Japan\",\"volume\":\"48 6\",\"pages\":\"0\"},\"PeriodicalIF\":2.2000,\"publicationDate\":\"2023-10-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Publications of the Astronomical Society of Japan\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1093/pasj/psad071\",\"RegionNum\":4,\"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}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Publications of the Astronomical Society of Japan","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1093/pasj/psad071","RegionNum":4,"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

摘要

蓝色水平分支恒星(BHBs)由于其明亮和几乎恒定的星等而成为研究银河系(MW)的理想示踪剂。然而,调查中对BHBs的不完整筛选会导致对MW结构或质量的估计存在偏差。通过像斯隆数字巡天(SDSS)这样的大型天文望远镜的巡天,有可能获得一个完整的样本。因此,快速有效地从海量测光图像中检测BHBs是必要的。目前bhb的采集方法主要是手动或半自动模式。因此,需要新的方法来取代人工或传统的机器学习检测。主流的基于深度学习的目标检测方法通常是普通的卷积神经网络,其提取全局特征的能力受到卷积算子接受域的限制。近年来,一种新的基于transformer的方法利用自注意机制带来的全局感受场优势,在许多任务中超越了普通卷积模型,并取得了优异的效果。因此,本文提出了一种混合卷积和Transformer模型,称为astrroyolo,以利用卷积在局部特征表示和Transformer更容易发现长距离特征依赖的优势。我们在4799 SDSS DR16光度图像数据集上进行了对比实验。实验结果表明,该模型在测试数据集上的准确率分别达到99.25% AP@50、93.79% AP@75和64.45% AP@95,优于YOLOv3和YOLOv4的目标检测模型。此外,我们在相同分辨率的基础上测试了更大的剪切图像。我们的模型分别可以达到99.02% AP@50、92.00% AP@75和61.96% AP@95,仍然优于YOLOv3和YOLOv4。这些结果还表明,适当的切割图像尺寸对于目标检测的性能和计算是必要的。与以往的模型相比,我们的模型取得了令人满意的目标检测结果,可以有效地提高BHB检测的精度。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
AstroYOLO: A hybrid CNN–Transformer deep-learning object-detection model for blue horizontal-branch stars
Abstract Blue horizontal-branch stars (BHBs) are ideal tracers for studying the Milky Way (MW) due to their bright and nearly constant magnitude. However, an incomplete screen of BHBs from a survey would result in bias of estimation of the structure or mass of the MW. With surveys of large sky telescopes like the Sloan Digital Sky Survey (SDSS), it is possible to obtain a complete sample. Thus, detecting BHBs from massive photometric images quickly and effectually is necessary. The current acquisition methods of BHBs are mainly based on manual or semi-automatic modes. Therefore, novel approaches are required to replace manual or traditional machine-learning detection. The mainstream deep-learning-based object-detection methods are often vanilla convolutional neural networks whose ability to extract global features is limited by the receptive field of the convolution operator. Recently, a new Transformer-based method has benefited from the global receptive field advantage brought by the self-attention mechanism, exceeded the vanilla convolution model in many tasks, and achieved excellent results. Therefore, this paper proposes a hybrid convolution and Transformer model called AstroYOLO to take advantage of the convolution in local feature representation and Transformer’s easier discovery of long-distance feature dependences. We conduct a comparative experiment on the 4799 SDSS DR16 photometric image dataset. The experimental results show that our model achieves 99.25% AP@50, 93.79% AP@75, and 64.45% AP@95 on the test dataset, outperforming the YOLOv3 and YOLOv4 object-detection models. In addition, we test on larger cutout images based on the same resolution. Our model can reach 99.02% AP@50, 92.00% AP@75, and 61.96% AP@95 respectively, still better than YOLOv3 and YOLOv4. These results also suggest that an appropriate size for cutout images is necessary for the performance and computation of object detection. Compared with the previous models, our model has achieved satisfactory object-detection results and can effectively improve the accuracy of BHB detection.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Publications of the Astronomical Society of Japan
Publications of the Astronomical Society of Japan 地学天文-天文与天体物理
CiteScore
4.10
自引率
13.00%
发文量
98
审稿时长
4-8 weeks
期刊介绍: Publications of the Astronomical Society of Japan (PASJ) publishes the results of original research in all aspects of astronomy, astrophysics, and fields closely related to them.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信