DeepStreaks: identifying fast-moving objects in the Zwicky Transient Facility data with deep learning - CNRS - Centre national de la recherche scientifique Accéder directement au contenu
Article Dans Une Revue Monthly Notices of the Royal Astronomical Society Année : 2019

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

Dmitry A. Duev
  • Fonction : Auteur
Ashish Mahabal
  • Fonction : Auteur
Quanzhi Ye
  • Fonction : Auteur
Kushal Tirumala
  • Fonction : Auteur
Justin Belicki
  • Fonction : Auteur
Richard Dekany
  • Fonction : Auteur
Sara Frederick
  • Fonction : Auteur
Matthew J. Graham
  • Fonction : Auteur
George Helou
  • Fonction : Auteur
Russ R. Laher
  • Fonction : Auteur
Frank J. Masci
  • Fonction : Auteur
Thomas A. Prince
  • Fonction : Auteur
Reed Riddle
  • Fonction : Auteur
Maayane T. Soumagnac
  • Fonction : Auteur

Résumé

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.

Dates et versions

hal-02153085 , version 1 (12-06-2019)

Identifiants

Citer

Dmitry A. Duev, Ashish Mahabal, Quanzhi Ye, Kushal Tirumala, Justin Belicki, et al.. DeepStreaks: identifying fast-moving objects in the Zwicky Transient Facility data with deep learning. Monthly Notices of the Royal Astronomical Society, 2019, 486 (3), pp.4158-4165. ⟨10.1093/mnras/stz1096⟩. ⟨hal-02153085⟩
33 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More