Radio Galaxy Zoo: CLARAN - A deep learning classifier for radio morphologies

Chen Wu, Oiwei Ivy Wong, Lawrence Rudnick, Stanislav S. Shabala, Matthew J. Alger, Julie K. Banfield, Cheng Soon Ong, Sarah V. White, Avery F. Garon, Ray P. Norris, Heinz Andernach, Jean Tate, Vesna Lukic, Hongming Tang, Kevin Schawinski, Foivos I. Diakogiannis

Research output: Contribution to journalArticlepeer-review

19 Scopus citations

Abstract

The upcoming next-generation large area radio continuum surveys can expect tens of millions of radio sources, rendering the traditional method for radio morphology classification through visual inspection unfeasible.We present CLARAN-Classifying Radio sources Automatically with Neural networks - a proof-of-concept radio source morphology classifier based upon the Faster Region-based Convolutional Neutral Networks method. Specifically, we train and test CLARAN on the FIRST and WISE (Wide-field Infrared Survey Explorer) images from the Radio Galaxy Zoo Data Release 1 catalogue. CLARAN provides end users with automated identification of radio source morphology classifications from a simple input of a radio image and a counterpart infrared image of the same region. CLARAN is the first open-source, endto- end radio source morphology classifier that is capable of locating and associating discrete and extended components of radio sources in a fast (<200 ms per image) and accurate (≥90 per cent) fashion. Future work will improve CLARAN's relatively lower success rates in dealing with multisource fields and will enable CLARAN to identify sources on much larger fields without loss in classification accuracy.

Original languageEnglish (US)
Pages (from-to)1211-1230
Number of pages20
JournalMonthly Notices of the Royal Astronomical Society
Volume482
Issue number1
DOIs
StatePublished - Jan 1 2019

Bibliographical note

Funding Information:
We acknowledge the contribution of more than 12 000 volunteers in the RGZ project. The International Centre for Radio Astronomy Research is a joint venture between Curtin University and The University of Western Australia with support and funding from the State Government of Western Australia. This work was supported by computing resources provided by the Pawsey Supercomputing Centre (with funding from the Australian Government and the Government of Western Australia) and CSIRO (Commonwealth Scientific and Industrial Research Organisation). Partial support for LR’s and AG’s work at the University of Minnesota comes from grant AST-1714205 from the U.S. National Science Foundation. HA benefited from grant DAIP no. 66/2018 of Universidad de Guanajuato.

Publisher Copyright:
© 2018 The Author(s). Published by Oxford University Press on behalf of the Royal Astronomical Society.

Keywords

  • Galaxies: active
  • Methods: numerical -methods: statistical
  • Radio continuum: galaxies
  • Techniques: image processing

Fingerprint

Dive into the research topics of 'Radio Galaxy Zoo: CLARAN - A deep learning classifier for radio morphologies'. Together they form a unique fingerprint.

Cite this