删除或更新信息,请邮件至freekaoyan#163.com(#换成@)

基于残差神经网络的恒星-星系分类器

本站小编 Free考研考试/2021-12-25

doi:10.12202/j.0476-0301.2021106杨阳1,
文中略2,
夏俊卿1,,
1.北京师范大学天文系,100875,北京
2.中国科学院国家天文台,100101,北京
基金项目:国家自然科学基金-天文联合基金重点资助项目(U1931202)

详细信息
通讯作者:夏俊卿(1983—),男,博士,教授. 研究方向:宇宙学的理论模型分析和实验数据处理. E-mail:xiajq@bnu.edu.cn
中图分类号:P152

计量

文章访问数:195
HTML全文浏览量:48
PDF下载量:21
被引次数:0
出版历程

收稿日期:2021-04-30
网络出版日期:2021-06-28
刊出日期:2021-09-02

Star-galaxy separation by the residual neural network algorithm

Yang YANG1,
Zhonglue WEN2,
Junqing XIA1,,
1. Department of Astronomy, Beijing Normal University, 100875, Beijing, China
2. National Astronomical Observatories, Chinese Academy of Sciences, 100101, Beijing, China



摘要
HTML全文
(12)(2)
参考文献(26)
相关文章
施引文献
资源附件(0)
访问统计

摘要
摘要:使用残差神经网络(residual neural network,RNN)算法对斯隆数字巡天(Sloan digital sky survey,SDSS)提供的天体伪彩色图片进行分类,直接从图像中获得特征.使用带有光谱信息的星系与恒星图片作为训练集和测试集.经过训练,在测试集上的准确率达到98.23%,召回率达到98.80%。这表明:RNN可以实现对星系和恒星图像的精确分类,分类器给出的恒星-星系概率是有效的,可用于分类可靠度评估;还可以尝试将此分类器应用到未来巡天中,进一步测试其性能.
关键词:巡天/
深度学习/
分类/
恒星/
星系
Abstract:In this paper, the residual neural network (RNN) algorithm was used to classify pseudo-color images of stars and galaxies from Sloan ?digital ?sky ?survey (SDSS), with features obtained directly from images.Images of galaxies and stars with spectral information were used as training and test sets.After training, accuracy rate on the test set can reach 98.23%, and recall rate 98.80%, indicating that the RNN algorithm can accurately classify images of galaxies and stars.The probability of being a star or galaxy given by the classifier is verified, and the probability can be used to evaluate the reliability of classification.This classifier can be applied to future sky surveys to further test its performance.
Key words:sky-survey/
deep learning/
classification/
star/
galaxy

相关话题/测试 北京 天文 图像 网络