NEO-BENCH: A New Multi-Source Benchmark for Generalizable Astronomical Streak Detection
Abstract
Near-Earth Objects (NEOs) can appear as faint streaks in long-exposure astronomical images. Detecting these streaks across diverse observatories requires methods that remain reliable despite differences in image quality, orientation, sky background, and noise. However, existing detectors are commonly evaluated using data from only one source, providing limited evidence of cross-source generalization. We introduce NEO-Bench, a multi-source benchmark containing 8,376 images from five astronomical-image datasets. The sources include the Hubble Space Telescope, a Stellina smart telescope, the United Arab Emirates Meteor Monitoring Network, a TETRA1 telescope using a Celestron C14 with Fastar, and the Roboflow Asteroid dataset. We converted the data to a common YOLO format, audited a sample of labels, and defined within-source and leave-one-source-out evaluation protocols. We evaluated four approaches: Hough, Radon, Gaussian PSF, and YOLO26L. Leave-one-source-out F1 decreased in 14 of 20 image-level method-source pairs and 13 of 20 [email protected] localization pairs. Across the datasets categorized as medium or hard, F1 decreased in 11 of 12 image-level pairs and 9 of 12 localization pairs. These results show that cross-source performance remains inconsistent and that reliable generalization across astronomical imaging sources remains an open challenge. The benchmark, code, and data are publicly available.
Related
- Dual-Phase Flow Matching for Probabilistic Source Detection and Measurement in Astronomical Images
- Fast and accurate astronomical source deblending with Density-Peak Clustering
- HLA-Net: A Physics-Guided Skeleton-Based Framework for Automatic Gravitational Arc Detection in Astronomical Images
- Misspecification in amortized X-ray spectral inference: a detection benchmark, a gain shift three schemes cannot see, and marginalizing it out
- ODIN: Comparison of Source Extractor and Tractor Source Detection for Lyman Alpha Emitting Galaxy Selection at z=3.1
BibTeX
@misc{he2026neobench,
title = {{NEO-BENCH: A New Multi-Source Benchmark for Generalizable Astronomical Streak Detection}},
author = {Jiayou He and Jessica Yao},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.06774},
eprint = {2609.06774},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.06774},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - NEO-BENCH: A New Multi-Source Benchmark for Generalizable Astronomical Streak Detection
AU - Jiayou He
AU - Jessica Yao
PY - 2026
DA - 2026/09/06
PB - arXiv
DO - 10.48550/arxiv.2609.06774
UR - https://arxiv.org/abs/2609.06774
AB - Near-Earth Objects (NEOs) can appear as faint streaks in long-exposure astronomical images. Detecting these streaks across diverse observatories requires methods that remain reliable despite differences in image quality, orientation, sky background, and noise. However, existing detectors are commonly evaluated using data from only one source, providing limited evidence of cross-source generalization. We introduce NEO-Bench, a multi-source benchmark containing 8,376 images from five astronomical-image datasets. The sources include the Hubble Space Telescope, a Stellina smart telescope, the United Arab Emirates Meteor Monitoring Network, a TETRA1 telescope using a Celestron C14 with Fastar, and the Roboflow Asteroid dataset. We converted the data to a common YOLO format, audited a sample of labels, and defined within-source and leave-one-source-out evaluation protocols. We evaluated four approaches: Hough, Radon, Gaussian PSF, and YOLO26L. Leave-one-source-out F1 decreased in 14 of 20 image-level method-source pairs and 13 of 20 [email protected] localization pairs. Across the datasets categorized as medium or hard, F1 decreased in 11 of 12 image-level pairs and 9 of 12 localization pairs. These results show that cross-source performance remains inconsistent and that reliable generalization across astronomical imaging sources remains an open challenge. The benchmark, code, and data are publicly available.
ER -