DeepDISC on LSST Data Preview 1: Scene-Level Detection, Deblending, and Star/Galaxy Classification in the ECDFS and EDFS Fields
Abstract
We present the first application of DeepDISC, a scene-level deep-learning framework, to real LSST data, using the Data Preview 1 (DP1) coadded images of the Extended Chandra Deep Field-South and the Euclid Deep Field-South fields. In a single pass, DeepDISC jointly detects, segments, and classifies sources as stars or galaxies, rather than treating these tasks as separate steps as in traditional pipelines. We warm-start from a model pretrained on LSST simulations and fine-tune it on DP1 six-band images, using the DP1 catalog's extendedness flag (point-like versus extended) as provisional class labels. The classification part of the network is then fine-tuned alone on a curated sample of stars and galaxies with spectroscopic, Gaia, and photometric identifications, so the final output is a star/galaxy classification. Fine-tuning on real DP1 observations substantially improves source recovery relative to the un-fine-tuned model, raising galaxy and star completeness from 40.2% and 13% to 84.2% and 49.1% under strict class-aware matching; under class-agnostic matching the fine-tuned model reaches 84.9% and 74.3%. On a common sample of sources that neither classifier saw in training, DeepDISC classifies to within about one percentage point of a published photometry-based Random Forest for the same field (97.4% vs. 98.5%); at fainter magnitudes DeepDISC labels more sources as stars than the Random Forest, and independent labels favor the Random Forest in most of these disagreements. Cross-matching to deeper HST/CANDELS imaging on a held-out footprint gives an unrecognized-blend fraction for DeepDISC of 14.9% (95% CI 12.5-17.8%), consistent with or modestly below the 18.0% we measure for the production DP1 catalog. This work establishes a baseline for DeepDISC on real LSST data and a foundation for self-supervised pretraining and for LSST Data Preview 2.
Related
- Rubin J122659.4+090236: An Extremely Low Surface Brightness Galaxy Candidate Discovered in the Rubin LSST Early Data Preview 2
- Galaxy clusters in the VIDEO fields: Detection and characterisation in the context of MOONRISE
- Fast and accurate astronomical source deblending with Density-Peak Clustering
- FOMO-Z. Optimal galaxy classification, photometric redshifts, and reconstruction of population distributions
- An Efficient Galaxy Morphology Classification Framework updated with Two-Stage Feature Extraction with AlexNet Encoding and UMAP
BibTeX
@misc{yadav2026deepdisc,
title = {{DeepDISC on LSST Data Preview 1: Scene-Level Detection, Deblending, and Star/Galaxy Classification in the ECDFS and EDFS Fields}},
author = {Prince Yadav and Yaswant Sai Ejjagiri and Xin Liu and Grant Merz and the LSST Dark Energy Science Collaboration},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.10986},
eprint = {2610.10986},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.10986},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - DeepDISC on LSST Data Preview 1: Scene-Level Detection, Deblending, and Star/Galaxy Classification in the ECDFS and EDFS Fields
AU - Prince Yadav
AU - Yaswant Sai Ejjagiri
AU - Xin Liu
AU - Grant Merz
AU - the LSST Dark Energy Science Collaboration
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.10986
UR - https://arxiv.org/abs/2610.10986
AB - We present the first application of DeepDISC, a scene-level deep-learning framework, to real LSST data, using the Data Preview 1 (DP1) coadded images of the Extended Chandra Deep Field-South and the Euclid Deep Field-South fields. In a single pass, DeepDISC jointly detects, segments, and classifies sources as stars or galaxies, rather than treating these tasks as separate steps as in traditional pipelines. We warm-start from a model pretrained on LSST simulations and fine-tune it on DP1 six-band images, using the DP1 catalog's extendedness flag (point-like versus extended) as provisional class labels. The classification part of the network is then fine-tuned alone on a curated sample of stars and galaxies with spectroscopic, Gaia, and photometric identifications, so the final output is a star/galaxy classification. Fine-tuning on real DP1 observations substantially improves source recovery relative to the un-fine-tuned model, raising galaxy and star completeness from 40.2% and 13% to 84.2% and 49.1% under strict class-aware matching; under class-agnostic matching the fine-tuned model reaches 84.9% and 74.3%. On a common sample of sources that neither classifier saw in training, DeepDISC classifies to within about one percentage point of a published photometry-based Random Forest for the same field (97.4% vs. 98.5%); at fainter magnitudes DeepDISC labels more sources as stars than the Random Forest, and independent labels favor the Random Forest in most of these disagreements. Cross-matching to deeper HST/CANDELS imaging on a held-out footprint gives an unrecognized-blend fraction for DeepDISC of 14.9% (95% CI 12.5-17.8%), consistent with or modestly below the 18.0% we measure for the production DP1 catalog. This work establishes a baseline for DeepDISC on real LSST data and a foundation for self-supervised pretraining and for LSST Data Preview 2.
ER -