DeepStreaks: identifying fast-moving objects in the Zwicky Transient Facility data with deep learning

被引:31
作者
Duev, Dmitry A. [1 ]
Mahabal, Ashish [1 ]
Ye, Quanzhi [1 ,2 ]
Tirumala, Kushal [3 ]
Belicki, Justin [4 ]
Dekany, Richard [4 ]
Frederick, Sara [5 ]
Graham, Matthew J. [1 ]
Laher, Russ R. [2 ]
Masci, Frank J. [2 ]
Prince, Thomas A. [1 ]
Riddle, Reed [4 ]
Rosnet, Philippe [6 ]
Soumagnac, Maayane T. [7 ]
机构
[1] CALTECH, Div Phys Math & Astron, Pasadena, CA 91125 USA
[2] CALTECH, IPAC, MS 100-22, Pasadena, CA 91125 USA
[3] CALTECH, Div Engn & Appl Sci, Pasadena, CA 91125 USA
[4] CALTECH, Caltech Opt Observ, Pasadena, CA 91125 USA
[5] Univ Maryland, Dept Astron, College Pk, MD 20742 USA
[6] Univ Clermont Auvergne, CNRS, IN2P3, LPC, F-63000 Clermont Ferrand, France
[7] Weizmann Inst Sci, Benoziyo Ctr Astrophys, IL-7610001 Rehovot, Israel
基金
美国国家科学基金会;
关键词
methods: data analysis; surveys; minor planets; asteroids:; general;
D O I
10.1093/mnras/stz1096
中图分类号
P1 [天文学];
学科分类号
0704 ;
摘要
We present DeepStreaks, a convolutional-neural-network, deep-learning system designed to efficiently identify streaking fast-moving near-Earth objects that are detected in the data of the Zwicky Transient Facility (ZTF), a wide-field, time-domain survey using a dedicated 47 deg(2) camera attached to the Samuel Oschin 48-inch Telescope at the Palomar Observatory in California, United States. The system demonstrates a 96-98 per cent true positive rate, depending on the night, while keeping the false positive rate below 1 per cent. The sensitivity of DeepStreaks is quantified by the performance on the test data sets as well as using known near-Earth objects observed by ZTF. The system is deployed and adapted for usage within the ZTF Solar system framework and has significantly reduced human involvement in the streak identification process, from several hours to typically under 10 min per day.
引用
收藏
页码:4158 / 4165
页数:8
相关论文
共 20 条
[1]  
Abadi M., 2016, 12 USENIX S OPERATIN, DOI 10.5555/3026877.3026899
[2]   The Zwicky Transient Facility: Surveys and Scheduler [J].
Bellm, Eric C. ;
Kulkarni, Shrinivas R. ;
Barlow, Tom ;
Feindt, Ulrich ;
Graham, Matthew J. ;
Goobar, Ariel ;
Kupfer, Thomas ;
Ngeow, Chow-Choong ;
Nugent, Peter ;
Ofek, Eran ;
Prince, Thomas A. ;
Riddle, Reed ;
Walters, Richard ;
Ye, Quan-Zhi .
PUBLICATIONS OF THE ASTRONOMICAL SOCIETY OF THE PACIFIC, 2019, 131 (1000)
[3]  
Chollet F., 2015, Keras
[4]  
Deng J, 2009, PROC CVPR IEEE, P248, DOI 10.1109/CVPRW.2009.5206848
[5]   Statistical and numerical study of asteroid orbital uncertainty [J].
Desmars, J. ;
Bancelin, D. ;
Hestroffer, D. ;
Thuillot, W. .
ASTRONOMY & ASTROPHYSICS, 2013, 554
[6]  
Graham M. J., 2019, PASP
[7]  
He K., 2015, P IEEE C COMPUTER VI, DOI DOI 10.1109/CVPR.2016.90
[8]  
HUANG G, 2017, PROC CVPR IEEE, P2261, DOI DOI 10.1109/CVPR.2017.243
[9]  
Jedicke R., 2013, ATLAS PS2 ASTEROID R
[10]  
Kingma D. P., 2014, arXiv, DOI DOI 10.48550/ARXIV.1412.6980