The Blending ToolKit: A simulation framework for evaluation of galaxy detection and deblending

• DOI: 10.33232/001c.129699
• Publication Date: 2025-02-14T07:29:21Z
• Authors: Ismael Mendoza; Andrii Torchylo; Thomas Sainrat; Axel Guinot; Alexandre Boucaud; Maxime Paillasa; Camille Avestruz; Prakruth Adari; Eric Aubourg; Biswajit Biswas; James Buchanan; Patricia Burchat; Cyrille Doux; Remy Joseph; Sowmya Kamath; Alex I. Malz; Grant Merz; Hironao Miyatake; Cécile Roucelle; Tianqing Zhang
• Keywords: QB460-466; Cosmology and Nongalactic Astrophysics (astro-ph.CO); Astronomy; FOS: Physical sciences; QB1-991; [PHYS.PHYS.PHYS-INS-DET]Physics [physics]/Physics [physics]/Instrumentation and Detectors [physics.ins-det]; [PHYS.ASTR]Physics [physics]/Astrophysics [astro-ph]; Astrophysics; Astrophysics - Instrumentation and Methods for Astrophysics; Instrumentation and Methods for Astrophysics (astro-ph.IM); 520; Astrophysics - Cosmology and Nongalactic Astrophysics

We present an open source Python library for simulating overlapping (i.e., blended) images of galaxies and performing self-consistent comparisons detection deblending algorithms based on a suite metrics. The package, named Blending Toolkit (BTK), serves as modular, flexible, easy-to-install, simple-to-use interface exploring analyzing systematic effects related to blended in cosmological surveys such the Vera Rubin Observatory Legacy Survey Space Time (LSST). BTK has three main components: (1) set modules that perform fast image simulations galaxies, using simulation package GalSim; (2) module standardizes inputs outputs existing algorithms; (3) metrics commonly defined galaxy literature. In combination, these allow researchers explore impacts blending surveys. Additionally, provides who are developing new algorithm framework evaluate performance make principled with deblenders. includes tutorials comprehensive documentation. code is publicly available GitHub at https://github.com/LSSTDESC/BlendingToolKit.