Radio Galaxy Zoo: host galaxies and radio morphologies for large surveys from visual inspection

TL;DR

Radio Galaxy Zoo leverages citizen science to classify radio sources and identify host galaxies, achieving expert-level accuracy and discovering rare objects, thereby supporting future large-scale radio surveys.

Contribution

This study demonstrates the effectiveness of citizen science in classifying complex radio morphologies and matching hosts, providing a large, validated catalog for future research and automation.

Findings

Over 1 million classifications for 60,000 sources

Volunteer classifications match expert accuracy at 75% consensus

Discovery of rare objects like HyMORs and giant radio galaxies

Abstract

We present early results from Radio Galaxy Zoo, a web-based citizen science project for visual inspection and classification of images from all-sky radio surveys. The goals of the project are to classify individual radio sources (particularly galaxies with multiple lobes and/or complex morphologies) as well as matching the continuum radio emission to the host galaxy. Radio images come from the FIRST and ATLAS surveys, while matches to potential hosts are performed with infrared imaging from WISE and SWIRE. The first twelve months of classification yielded more than 1 million classifications of more than 60,000 sources. For images with at least 75% consensus by the volunteer classifiers, the accuracy is comparable to visual inspection by the expert science team. Based on mid-infrared colors, the hosts associated with radio emission are primarily a mixture of elliptical galaxies, QSOs, and LIRGs, which are in good agreement with previous studies. The full catalog of radio lobes and their host galaxies will measure the relative populations of host types as a function of radio morphology and power. Radio Galaxy Zoo has also been an effective method for detecting extremely rare objects, such as HyMORs and giant radio galaxies. Results from the project are being used to train automatic algorithms for host matching for use in future large continuum surveys such as EMU, as well as establishing roles for citizen science in projects such as the SKA.