Identification of gravitational lenses obscured by foreground light in the KiDS dataset using U-Nets and ResNets
Abstract
*Context.* Many lensing images are often obscured by foreground light from the central galaxies, making them challenging to detect. *Aims.* To address the limitations of previous lens search efforts, particularly for samples with smaller RE or faint lensed images, we developed a composite convolutional neural network framework that utilizes both U-Net and ResNet architectures for feature extraction and classification. *Methods.* We propose a hybrid search method that combines U-Net and ResNet architectures to enhance the detection of foreground galaxy-obscured lenses. Our approach consists of two main stages: first, the U-Net model separates the foreground galaxy light from potential lensing signals, creating residual images that highlight the lensing features. Next, the ResNet module performs binary classification on these residual images to detect lensing signals. *Results.* We evaluated the hybrid search method with real observational data to demonstrate its effectiveness, achieving a recall of 71.5% and a 4.5% false positive rate at a confidence threshold of 0.6. Applying this method to over 638,398 galaxy samples from the Kilo-Degree Survey Data Release 4 and conducting thorough inspections, we identify 88 Class A, 322 Class B, and 1,758 Class C candidates. *Conclusions.* This hybrid approach significantly enhances the completeness of existing strong gravitational lensing searches and shows great potential for improving future astronomical surveys.
Versions
Related
- SPT-3G D1: Foreground-Robust Lensing Templates for Primordial Gravitational Wave Searches
- Automated identification and subtraction of gravitational-wave glitches using boundary refinement
- Fisher Information Velocity: A New Geometric Channel for Precision Glitch Identification in Gravitational-Wave Detectors
- The obscured nucleus of NGC 3256,S. I. Anisotropic active galactic nucleus near-IR radiation field traced by CO v=1--0 with JWST NIRSpec
- JESSNet: Joint dEconvolution and Sparse Separation Network for HI 21 cm foreground cleaning
BibTeX
@misc{liu2026identification,
title = {{Identification of gravitational lenses obscured by foreground light in the KiDS dataset using U-Nets and ResNets}},
author = {S. Liu and Rui Li and J. Jia and Hui Li and Liqing Chen and Xiaoyue Cao and Zizhao He and Valerio Busillo and Nicola N. Napolitano and Crescenzo Tortora and Fucheng Zhong and Hao Su and Haicheng Feng and Yue Dong and Ran Li and Liang Gao},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.21660},
eprint = {2609.21660},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.21660},
}
RIS
TY - UNPB
TI - Identification of gravitational lenses obscured by foreground light in the KiDS dataset using U-Nets and ResNets
AU - S. Liu
AU - Rui Li
AU - J. Jia
AU - Hui Li
AU - Liqing Chen
AU - Xiaoyue Cao
AU - Zizhao He
AU - Valerio Busillo
AU - Nicola N. Napolitano
AU - Crescenzo Tortora
AU - Fucheng Zhong
AU - Hao Su
AU - Haicheng Feng
AU - Yue Dong
AU - Ran Li
AU - Liang Gao
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.21660
UR - https://arxiv.org/abs/2609.21660
AB - *Context.* Many lensing images are often obscured by foreground light from the central galaxies, making them challenging to detect. *Aims.* To address the limitations of previous lens search efforts, particularly for samples with smaller R_E or faint lensed images, we developed a composite convolutional neural network framework that utilizes both U-Net and ResNet architectures for feature extraction and classification. *Methods.* We propose a hybrid search method that combines U-Net and ResNet architectures to enhance the detection of foreground galaxy-obscured lenses. Our approach consists of two main stages: first, the U-Net model separates the foreground galaxy light from potential lensing signals, creating residual images that highlight the lensing features. Next, the ResNet module performs binary classification on these residual images to detect lensing signals. *Results.* We evaluated the hybrid search method with real observational data to demonstrate its effectiveness, achieving a recall of 71.5% and a 4.5% false positive rate at a confidence threshold of 0.6. Applying this method to over 638,398 galaxy samples from the Kilo-Degree Survey Data Release 4 and conducting thorough inspections, we identify 88 Class A, 322 Class B, and 1,758 Class C candidates. *Conclusions.* This hybrid approach significantly enhances the completeness of existing strong gravitational lensing searches and shows great potential for improving future astronomical surveys.
ER -