LUMA: A CNN for Strong Gravitational Lens Searches in Astronomical Imaging
Abstract
We present LUMA, a convolutional neural network (CNN) pipeline for the automated detection of strong gravitational lenses in simulated astronomical imaging. The method combines a physically motivated preprocessing stage, which enhances faint arc and ring features, with a compact three-block CNN trained using class reweighting and modern learning-rate scheduling. On simulated data, the model reaches test accuracies of about $96$\% and receiver operating characteristic (ROC) area-under-the-curve (AUC) values of ≃ 0.99 for the non-trivial classes, while confusion-matrix analysis shows high completeness and purity for lens candidates. These results demonstrate that relatively lightweight CNN architectures can provide a competitive baseline for strong-lens searches, and they motivate future extensions toward real survey images and transformer-based models.
Versions
Related
- Stochastic effect on ultralight dark matter searches with astronomical polarized sources
- HLA-Net: A Physics-Guided Skeleton-Based Framework for Automatic Gravitational Arc Detection in Astronomical Images
- A semicoherent resampling method for long-transient gravitational wave searches with applications to subsolar-mass primordial black-hole binaries
- Neural Network Guided Parameter Space Constraints for Gravitational Wave Searches from Binary Black Holes
- Alternative neural-network follow-up for all-sky FrequencyHough in continuous gravitational-wave searches
BibTeX
@misc{donatiello2026luma,
title = {{LUMA: A CNN for Strong Gravitational Lens Searches in Astronomical Imaging}},
author = {Giovanni Vincenzo Donatiello and Achille A. Nucita and Francesco De Paolis and Antonio Franco and Francesco Strafella},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.36857},
eprint = {2609.36857},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.36857},
}
RIS
TY - UNPB
TI - LUMA: A CNN for Strong Gravitational Lens Searches in Astronomical Imaging
AU - Giovanni Vincenzo Donatiello
AU - Achille A. Nucita
AU - Francesco De Paolis
AU - Antonio Franco
AU - Francesco Strafella
PY - 2026
DA - 2026/09/29
PB - arXiv
DO - 10.48550/arxiv.2609.36857
UR - https://arxiv.org/abs/2609.36857
AB - We present LUMA, a convolutional neural network (CNN) pipeline for the automated detection of strong gravitational lenses in simulated astronomical imaging. The method combines a physically motivated preprocessing stage, which enhances faint arc and ring features, with a compact three-block CNN trained using class reweighting and modern learning-rate scheduling. On simulated data, the model reaches test accuracies of about $96$\% and receiver operating characteristic (ROC) area-under-the-curve (AUC) values of ≃ 0.99 for the non-trivial classes, while confusion-matrix analysis shows high completeness and purity for lens candidates. These results demonstrate that relatively lightweight CNN architectures can provide a competitive baseline for strong-lens searches, and they motivate future extensions toward real survey images and transformer-based models.
ER -