A Catalog of Visual-like Morphologies in the 5 CANDELS Fields Using Deep Learning - CNRS - Centre national de la recherche scientifique Accéder directement au contenu
Article Dans Une Revue The Astrophysical Journal Supplement Année : 2015

A Catalog of Visual-like Morphologies in the 5 CANDELS Fields Using Deep Learning

G. Cabrera-Vives
  • Fonction : Auteur
P. Pérez-González
J. Kartaltepe
G. Barro
  • Fonction : Auteur
F. Shankar
E. Bell
  • Fonction : Auteur
D. Kocevski
  • Fonction : Auteur
D. Koo
  • Fonction : Auteur
S. Faber
  • Fonction : Auteur
D. Mcintosh
  • Fonction : Auteur

Résumé

We present a catalog of visual-like H-band morphologies of ̃50.000 galaxies (Hf160w < 24.5) in the 5 CANDELS fields (GOODS-N, GOODS-S, UDS, EGS, and COSMOS). Morphologies are estimated using Convolutional Neural Networks (ConvNets). The median redshift of the sample is < z> ̃ 1.25. The algorithm is trained on GOODS-S, for which visual classifications are publicly available, and then applied to the other 4 fields. Following the CANDELS main morphology classification scheme, our model retrieves for each galaxy the probabilities of having a spheroid or a disk, presenting an irregularity, being compact or a point source, and being unclassifiable. ConvNets are able to predict the fractions of votes given to a galaxy image with zero bias and ̃10% scatter. The fraction of mis-classifications is less than 1%. Our classification scheme represents a major improvement with respect to Concentration-Asymmetry-Smoothness-based methods, which hit a 20%-30% contamination limit at high z. The catalog is released with the present paper via the Rainbow database (http://rainbowx.fis.ucm.es/Rainbow_navigator_public/).

Dates et versions

hal-02453326 , version 1 (23-01-2020)

Identifiants

Citer

M. Huertas-Company, R. Gravet, G. Cabrera-Vives, P. Pérez-González, J. Kartaltepe, et al.. A Catalog of Visual-like Morphologies in the 5 CANDELS Fields Using Deep Learning. The Astrophysical Journal Supplement, 2015, 221 (1), pp.8. ⟨10.1088/0067-0049/221/1/8⟩. ⟨hal-02453326⟩
21 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More