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 journalArticle

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, end-to-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
Pages (from-to)1211-1230
Number of pages20
JournalMonthly Notices of the Royal Astronomical Society
Volume482
Issue number1
DOIs
Publication statusPublished - Jan 2019

Cite this

Wu, Chen ; Wong, Oiwei Ivy ; Rudnick, Lawrence ; Shabala, Stanislav S. ; Alger, Matthew J. ; Banfield, Julie K. ; Ong, Cheng Soon ; White, Sarah V. ; Garon, Avery F. ; Norris, Ray P. ; Andernach, Heinz ; Tate, Jean ; Lukic, Vesna ; Tang, Hongming ; Schawinski, Kevin ; Diakogiannis, Foivos I. / Radio Galaxy Zoo : Claran - a deep learning classifier for radio morphologies. In: Monthly Notices of the Royal Astronomical Society. 2019 ; Vol. 482, No. 1. pp. 1211-1230.
@article{fda65eba08944754abc00fcc0bc71563,
title = "Radio Galaxy Zoo: Claran - a deep learning classifier for radio morphologies",
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, end-to-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.",
keywords = "methods: numerical, methods: statistical, techniques: image processing, galaxies: active, radio continuum: galaxies, CONVOLUTIONAL NEURAL-NETWORKS, EXOPLANETS, COMPACT, 1ST, SKY",
author = "Chen Wu and Wong, {Oiwei Ivy} and Lawrence Rudnick and Shabala, {Stanislav S.} and Alger, {Matthew J.} and Banfield, {Julie K.} and Ong, {Cheng Soon} and White, {Sarah V.} and Garon, {Avery F.} and Norris, {Ray P.} and Heinz Andernach and Jean Tate and Vesna Lukic and Hongming Tang and Kevin Schawinski and Diakogiannis, {Foivos I.}",
year = "2019",
month = "1",
doi = "10.1093/mnras/sty2646",
language = "English",
volume = "482",
pages = "1211--1230",
journal = "Monthly Notices of the Royal Astronomical Society",
issn = "0035-8711",
publisher = "OXFORD UNIV PRESS UNITED KINGDOM",
number = "1",

}

Wu, C, Wong, OI, Rudnick, L, Shabala, SS, Alger, MJ, Banfield, JK, Ong, CS, White, SV, Garon, AF, Norris, RP, Andernach, H, Tate, J, Lukic, V, Tang, H, Schawinski, K & Diakogiannis, FI 2019, 'Radio Galaxy Zoo: Claran - a deep learning classifier for radio morphologies' Monthly Notices of the Royal Astronomical Society, vol. 482, no. 1, pp. 1211-1230. https://doi.org/10.1093/mnras/sty2646

Radio Galaxy Zoo : Claran - a deep learning classifier for radio morphologies. / Wu, Chen; Wong, Oiwei Ivy; Rudnick, Lawrence; Shabala, Stanislav S.; Alger, Matthew J.; Banfield, Julie K.; Ong, Cheng Soon; White, Sarah V.; Garon, Avery F.; Norris, Ray P.; Andernach, Heinz; Tate, Jean; Lukic, Vesna; Tang, Hongming; Schawinski, Kevin; Diakogiannis, Foivos I.

In: Monthly Notices of the Royal Astronomical Society, Vol. 482, No. 1, 01.2019, p. 1211-1230.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Radio Galaxy Zoo

T2 - Claran - a deep learning classifier for radio morphologies

AU - Wu, Chen

AU - Wong, Oiwei Ivy

AU - Rudnick, Lawrence

AU - Shabala, Stanislav S.

AU - Alger, Matthew J.

AU - Banfield, Julie K.

AU - Ong, Cheng Soon

AU - White, Sarah V.

AU - Garon, Avery F.

AU - Norris, Ray P.

AU - Andernach, Heinz

AU - Tate, Jean

AU - Lukic, Vesna

AU - Tang, Hongming

AU - Schawinski, Kevin

AU - Diakogiannis, Foivos I.

PY - 2019/1

Y1 - 2019/1

N2 - 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, end-to-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.

AB - 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, end-to-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.

KW - methods: numerical

KW - methods: statistical

KW - techniques: image processing

KW - galaxies: active

KW - radio continuum: galaxies

KW - CONVOLUTIONAL NEURAL-NETWORKS

KW - EXOPLANETS

KW - COMPACT

KW - 1ST

KW - SKY

U2 - 10.1093/mnras/sty2646

DO - 10.1093/mnras/sty2646

M3 - Article

VL - 482

SP - 1211

EP - 1230

JO - Monthly Notices of the Royal Astronomical Society

JF - Monthly Notices of the Royal Astronomical Society

SN - 0035-8711

IS - 1

ER -