arrow
返回

Galaxy morphological classification using Dynamic Multiscale Attention Network

delete2025-08-01
delete0
delete
OA
AI
M
Ma, Baisen
B
Bo Qiu *
A
A-Li Luo *
Q
Qi Li *
F
Fuji Ren
M
Mengyao Li
DOI:10.1093/mnras/staf1037delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
大规模数字巡天计划的实施收集了海量的光度图像,使星系形态分类成为当前的研究热点。本文提出了一种基于动态大卷积核和注意力特征融合的星系形态分类网络,命名为Dynamic Multiscale Attention Galaxy Network (DMAGNet)。本研究使用了来自Galaxy Zoo DECaLS(暗能量相机遗产巡天)项目的15 266张星系图像,包含六类星系形态。这六类形态分别是edge-on、cigar、in-between、round、spiral和merger。在测试集上,该模型展现出卓越的性能,准确率达到97.1%,召回率为96.8%,F1分数为96.8%。此外,所提出的网络接受了广泛的评估和消融实验,并采用t-分布随机邻域嵌入(t-SNE)方法可视化测试集中提取的形态特征。Maxvit的准确率为0.969,而DMAGNet的准确率达到0.971。实验结果表明,该模型在星系形态分类任务中优于已知的分类模型。最后,该方法被应用于DESI遗产巡天BASS + MzLS数据集中的14 174 190张星系图像,构建了一个覆盖六类形态的星系目录:Edge-On(192 577)、In-Between(2353 937)、Cigar(299 725)、Round(353 358)、Spiral(16 067)和Merger(1266 961)。此外,引入了一个Error列,包含总计9691 564个样本,通过整合其他目录的信息来标记可能损坏或存在质量问题的图像。
Keyword:
methods: data analysis
techniques: image processing
Galaxies: structure

期刊

Monthly Notices of the Royal Astronomical Society 封面图
Monthly Notices of the Royal Astronomical Society
IF:
4.8
论文数:
7.0W
被引数:
25.0W

机构

H
Hebei University of Technology
学者数:
3.5K
论文数: 1.0K
被引数: 1.7W
C
Chinese Academy of Sciences
学者数:
3.9W
论文数: 1.5W
被引数: 58.4W
引用论文

引用论文

Galaxy Zoo: comparing the demographics of spiral arm number and a new method for correcting redshift bias
err2016-07-04
err103
errOAAI
errHart, Ross E.; Bamford, Steven P.; Willett, Kyle W.; Masters, Karen L.; Cardamone, Carolin; Lintott, Chris J.; Mackay, Robert J.; Nichol, Robert C.; Rosslowe, Christopher K.; Simmons, Brooke D.; Smethurst, Rebecca J.
err分享
err收藏
Galaxy Zoo DECaLS: Detailed visual morphology measurements from volunteers and deep learning for 314 000 galaxies
err2021-09-30
err85
errOAAI
errWalmsley, Mike; Lintott, Chris; Geron, Tobias; Kruk, Sandor; Krawczyk, Coleman; Willett, Kyle W.; Bamford, Steven; Kelvin, Lee S.; Fortson, Lucy; Gal, Yarin; Keel, William; Masters, Karen L.; Mehta, Vihang; Simmons, Brooke D.; Smethurst, Rebecca; Smith, Lewis; Baeten, Elisabeth M.; Macmillan, Christine
err分享
err收藏
Automatic classification of galaxy morphology based on the RegNetX-CBAM3 algorithm
err2022-09-26
err4
PREAI
errLi, Juan; Tu, Liangping; Gao, Xiang; Li, Xin; Zhong, Zhengdi; Feng, Xueqi
err分享
err收藏
Single Image Super-Resolution via a Holistic Attention Network
err2020-10-07
err0
PREAI
errBen Niu; Weilei Wen; Wenqi Ren; Xiangde Zhang; Lianping Yang; Shuzhen Wang; Kaihao Zhang; Xiaochun Cao; Haifeng Shen
err分享
err收藏
err
IF0
err
err0
errOAAI
err
err分享
err收藏
Galaxy morphology classification with deep convolutional neural networks
err2019-04-02
err0
errOAAI
errXiao-Pan Zhu; Jia-Ming Dai; Chun-Jiang Bian; Yu Chen; Shi Chen; Chen Hu
err分享
err收藏
学者 查看更多内容