Strong lensing in UNIONS: Toward a pipeline from discovery to modeling

TL;DR

This paper develops an end-to-end pipeline using CNNs and modeling techniques to discover and analyze galaxy-scale strong gravitational lenses in large survey data, demonstrating its effectiveness on CFIS data and preparing for future Euclid surveys.

Contribution

It introduces the first integrated pipeline combining CNN-based detection, visual inspection, automated modeling, and deblending for strong lensing in a single band survey.

Findings

Identified 133 strong lens candidates, including 104 new ones.

Successfully modeled 32 high-quality lenses to derive Einstein radii.

Developed a novel auto-encoder based deblending algorithm for lens/source separation.

Abstract

We present a search for galaxy-scale strong gravitational lenses in the initial 2 500 square degrees of the Canada-France Imaging Survey (CFIS). We designed a convolutional neural network (CNN) committee that we applied to a selection of 2 344 002 exquisite-seeing $r$-band images of color-selected luminous red galaxies (LRGs). Our classification uses a realistic training set where the lensing galaxies and the lensed sources are both taken from real data, namely the CFIS $r$-band images themselves and the Hubble Space Telescope (HST). A total of 9 460 candidates obtain a score above 0.5 with the CNN committee. After a visual inspection of the candidates, we find a total of 133 lens candidates, of which 104 are completely new. The set of false positives mainly contains ring, spiral, and merger galaxies, and to a lesser extent galaxies with nearby companions. We classify 32 of the lens candidates as secure lenses and 101 as maybe lenses. For the 32 highest quality lenses, we also fit a singular isothermal ellipsoid mass profile with external shear along with an elliptical Sersic profile for the lens and source light. This automated modeling step provides distributions of properties for both sources and lenses that have Einstein radii in the range $0.5\arcsec<\theta_E<2.5\arcsec$. Finally, we introduce a new lens and/or source single-band deblending algorithm based on auto-encoder representation of our candidates. This is the first time an end-to-end lens-finding and modeling pipeline is assembled together, in view of future lens searches in a single band, as will be possible with Euclid.