Plate-YOLO: An End-to-End Object Detection Framework for Robust Astrometric Registration of Historical Plates
Abstract
Historical astronomical plates provide a long temporal baseline for time-domain astrophysics, but degradation and structural artifacts can hinder automated astrometric calibration. We present Plate-YOLO, a framework that combines source detection and stellar classification in one detector and supplies source catalogs to a separate astrometric solver. The detector adapts YOLOv5 with Space-to-Depth feature rearrangement, SimAM attention, a high-resolution P2 head, and a Normalized Wasserstein Distance loss for small bounding boxes. A Gaia-guided fine-tuning procedure uses initially solved plates to construct labels for a new archive. On 150 UBAI plates for which the conventional pipeline had failed, the Chinese-trained model solved 40 plates, and target-archive fine-tuning recovered a further 106, giving an overall recovery rate of 146/150 (97.3%). In the reported hardware configuration, source extraction took 6.61 s per plate for Plate-YOLO and 16.73 s for SExtractor. These results demonstrate the practical value of adapting source detection to degraded historical plates.
Related
- Identifying Compact Object Candidates using Gaia DR3 Astrometric Accelerations and Jerks
- Beyond what's observed: hierarchical inference of Gaia astrometric compact-object binaries
- Astrometric detection of rocky exoplanets from inexpensive space platforms
- First Experimental Results of Space Object Detection Using the InteRFeX Array
- The JWST Proto-PAH Project: Detection of the methyl radical CH3 in the highly evolved C-rich object SMP LMC 011
BibTeX
@misc{xu2026plateyolo,
title = {{Plate-YOLO: An End-to-End Object Detection Framework for Robust Astrometric Registration of Historical Plates}},
author = {Quanfeng Xu and Zhengjun Shang and Shiyin Shen and Yong Yu and Mingyang Ma and Zhenghong Tang and Meiting Yang and Jing Yang and Jianhai Zhao},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.25677},
eprint = {2610.25677},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.25677},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Plate-YOLO: An End-to-End Object Detection Framework for Robust Astrometric Registration of Historical Plates
AU - Quanfeng Xu
AU - Zhengjun Shang
AU - Shiyin Shen
AU - Yong Yu
AU - Mingyang Ma
AU - Zhenghong Tang
AU - Meiting Yang
AU - Jing Yang
AU - Jianhai Zhao
PY - 2026
DA - 2026/10/19
PB - arXiv
DO - 10.48550/arxiv.2610.25677
UR - https://arxiv.org/abs/2610.25677
AB - Historical astronomical plates provide a long temporal baseline for time-domain astrophysics, but degradation and structural artifacts can hinder automated astrometric calibration. We present Plate-YOLO, a framework that combines source detection and stellar classification in one detector and supplies source catalogs to a separate astrometric solver. The detector adapts YOLOv5 with Space-to-Depth feature rearrangement, SimAM attention, a high-resolution P2 head, and a Normalized Wasserstein Distance loss for small bounding boxes. A Gaia-guided fine-tuning procedure uses initially solved plates to construct labels for a new archive. On 150 UBAI plates for which the conventional pipeline had failed, the Chinese-trained model solved 40 plates, and target-archive fine-tuning recovered a further 106, giving an overall recovery rate of 146/150 (97.3%). In the reported hardware configuration, source extraction took 6.61 s per plate for Plate-YOLO and 16.73 s for SExtractor. These results demonstrate the practical value of adapting source detection to degraded historical plates.
ER -