LinKS: discovering galaxy-scale strong lenses in the Kilo-Degree Survey using convolutional neural networks

Enrico Petrillo*, Crescenzo Tortora, Giorgos Vernardos, L.V.E. Koopmans, Gijs Verdoes Kleijn, M. Bilicki, Nicola Napolitano, Saikat Chatterjee, G Covone, Andrej Dvornik, Thomas Erben, Fedor Getman, B. Giblin, Catherine Heymans, J. T. A. de Jong, Koenraad Kuijken, Peter Schneider, Shan Huang, Chiara Spiniello, A. H. Wright

*Bijbehorende auteur voor dit werk

OnderzoeksoutputAcademicpeer review

45 Citaten (Scopus)
186 Downloads (Pure)


We present a new sample of galaxy-scale strong gravitational lens candidates, selected from 904 deg2 of Data Release 4 of the Kilo-Degree Survey, i.e. the ‘Lenses in the Kilo-Degree Survey’ (LinKS) sample. We apply two convolutional neural networks (ConvNets) to ∼88000 colour–magnitude-selected luminous red galaxies yielding a list of 3500 strong lens candidates. This list is further downselected via human inspection. The resulting LinKS sample is composed of 1983 rank-ordered targets classified as ‘potential lens candidates’ by at least one inspector. Of these, a high-grade subsample of 89 targets is identified with potential strong lenses by all inspectors. Additionally, we present a collection of another 200 strong lens candidates discovered serendipitously from various previous ConvNet runs. A straightforward application of our procedure to future Euclid or Large Synoptic Survey Telescope data can select a sample of ∼3000 lens candidates with less than 10 per cent expected false positives and requiring minimal human intervention.
Originele taal-2English
Pagina's (van-tot)3879–3896
Aantal pagina's18
TijdschriftMonthly Notices of the Royal Astronomical Society
Nummer van het tijdschrift3
StatusPublished - apr.-2019

Citeer dit