Streamlined lensed quasar identification in multiband images via ensemble networks

¹ Technical University of Munich, TUM School of Natural Sciences, Department of Physics, James-Franck-Str. 1, 85748 Garching, Germany
e-mail: irham.andika@tum.de
² Max-Planck-Institut für Astrophysik, Karl-Schwarzschild-Str. 1, 85748 Garching, Germany
³ Academia Sinica Institute of Astronomy and Astrophysics (ASIAA), 11F of ASMAB, No.1, Section 4, Roosevelt Road, Taipei 10617, Taiwan
⁴ Dipartimento di Fisica, Università degli Studi di Milano, via Celoria 16, 20133 Milano, Italy
⁵ Purple Mountain Observatory, No. 10 Yuanhua Road, Nanjing, Jiangsu, 210033, PR China
⁶ MIT Kavli Institute for Astrophysics and Space Research, 77 Massachusetts Ave., Cambridge, MA 02139, USA
⁷ Astronomy Research Group and Bosscha Observatory, FMIPA, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia
⁸ U-CoE AI-VLB, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia

Received: 30 June 2023
Accepted: 10 August 2023

Abstract

Quasars experiencing strong lensing offer unique viewpoints on subjects related to the cosmic expansion rate, the dark matter profile within the foreground deflectors, and the quasar host galaxies. Unfortunately, identifying them in astronomical images is challenging since they are overwhelmed by the abundance of non-lenses. To address this, we have developed a novel approach by ensembling cutting-edge convolutional networks (CNNs) - for instance, ResNet, Inception, NASNet, MobileNet, EfficientNet, and RegNet – along with vision transformers (ViTs) trained on realistic galaxy-quasar lens simulations based on the Hyper Suprime-Cam (HSC) multiband images. While the individual model exhibits remarkable performance when evaluated against the test dataset, achieving an area under the receiver operating characteristic curve of >97.3% and a median false positive rate of 3.6%, it struggles to generalize in real data, indicated by numerous spurious sources picked by each classifier. A significant improvement is achieved by averaging these CNNs and ViTs, resulting in the impurities being downsized by factors up to 50. Subsequently, combining the HSC images with the UKIRT, VISTA, and unWISE data, we retrieve approximately 60 million sources as parent samples and reduce this to 892 609 after employing a photometry preselection to discover z > 1.5 lensed quasars with Einstein radii of θ_E < 5″. Afterward, the ensemble classifier indicates 3080 sources with a high probability of being lenses, for which we visually inspect, yielding 210 prevailing candidates awaiting spectroscopic confirmation. These outcomes suggest that automated deep learning pipelines hold great potential in effectively detecting strong lenses in vast datasets with minimal manual visual inspection involved.