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

Dmitry A. Duev; Ashish Mahabal; Quanzhi Ye; Kushal Tirumala; Justin Belicki; Richard Dekany; Sara Frederick; Matthew J. Graham; Russ R. Laher; Frank J. Masci; Thomas A. Prince; Reed Riddle; Philippe Rosnet; Maayane T. Soumagnac

doi:10.1093/mnras/stz1096

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

Dmitry A. Duev, Ashish Mahabal, Quanzhi Ye, Kushal Tirumala, Justin Belicki, Richard Dekany, Sara Frederick, Matthew J. Graham, Russ R. Laher, Frank J. Masci, Thomas A. Prince, Reed Riddle, Philippe Rosnet, Maayane T. Soumagnac

Department of Particle Physics and Astrophysics

Weizmann Institute of Science

Research output: Contribution to journal › Article › peer-review

Abstract

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 deg² 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.

Original language	English
Pages (from-to)	4158-4165
Number of pages	8
Journal	MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY
Volume	486
Issue number	3
Early online date	17 Apr 2019
DOIs	https://doi.org/10.1093/mnras/stz1096
State	Published - 1 Jul 2019

Keywords

asteroids: general
methods: data analysis
minor planets
surveys

All Science Journal Classification (ASJC) codes

Astronomy and Astrophysics
Space and Planetary Science

Access to Document

10.1093/mnras/stz1096

https://arxiv.org/abs/1904.05920License: Other

Cite this

Duev, D. A., Mahabal, A., Ye, Q., Tirumala, K., Belicki, J., Dekany, R., Frederick, S., Graham, M. J., Laher, R. R., Masci, F. J., Prince, T. A., Riddle, R., Rosnet, P., & Soumagnac, M. T. (2019). DeepStreaks: Identifying fast-moving objects in the Zwicky Transient Facility data with deep learning. MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY, 486(3), 4158-4165. https://doi.org/10.1093/mnras/stz1096

@article{dcc0ef057ec44398aee55ddcf8ca64d2,

title = "DeepStreaks: Identifying fast-moving objects in the Zwicky Transient Facility data with deep learning",

abstract = "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 deg2 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.",

keywords = "asteroids: general, methods: data analysis, minor planets, surveys",

author = "Duev, {Dmitry A.} and Ashish Mahabal and Quanzhi Ye and Kushal Tirumala and Justin Belicki and Richard Dekany and Sara Frederick and Graham, {Matthew J.} and Laher, {Russ R.} and Masci, {Frank J.} and Prince, {Thomas A.} and Reed Riddle and Philippe Rosnet and Soumagnac, {Maayane T.}",

note = "Publisher Copyright: {\textcopyright} 2019 The Author(s) Published by Oxford University Press on behalf of the Royal Astronomical Society.",

year = "2019",

month = jul,

day = "1",

doi = "10.1093/mnras/stz1096",

language = "الإنجليزيّة",

volume = "486",

pages = "4158--4165",

journal = "MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY",

issn = "0035-8711",

publisher = "Oxford University Press",

number = "3",

}

Duev, DA, Mahabal, A, Ye, Q, Tirumala, K, Belicki, J, Dekany, R, Frederick, S, Graham, MJ, Laher, RR, Masci, FJ, Prince, TA, Riddle, R, Rosnet, P & Soumagnac, MT 2019, 'DeepStreaks: Identifying fast-moving objects in the Zwicky Transient Facility data with deep learning', MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY, vol. 486, no. 3, pp. 4158-4165. https://doi.org/10.1093/mnras/stz1096

TY - JOUR

T1 - DeepStreaks

T2 - Identifying fast-moving objects in the Zwicky Transient Facility data with deep learning

AU - Duev, Dmitry A.

AU - Mahabal, Ashish

AU - Ye, Quanzhi

AU - Tirumala, Kushal

AU - Belicki, Justin

AU - Dekany, Richard

AU - Frederick, Sara

AU - Graham, Matthew J.

AU - Laher, Russ R.

AU - Masci, Frank J.

AU - Prince, Thomas A.

AU - Riddle, Reed

AU - Rosnet, Philippe

AU - Soumagnac, Maayane T.

PY - 2019/7/1

Y1 - 2019/7/1

N2 - 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 deg2 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.

AB - 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 deg2 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.

KW - asteroids: general

KW - methods: data analysis

KW - minor planets

KW - surveys

UR - http://www.scopus.com/inward/record.url?scp=85070693490&partnerID=8YFLogxK

U2 - 10.1093/mnras/stz1096

DO - 10.1093/mnras/stz1096

M3 - مقالة

SN - 0035-8711

VL - 486

SP - 4158

EP - 4165

JO - MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY

JF - MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY

IS - 3

ER -

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

Abstract

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this