Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 13, 2026Astronomy and ComputingOpen Access

Aperture-X: Physics-informed aperture feature learning for robust photometry

View Full Paper
Ask AI
Bookmark
Share

Authors

KAKimmy De AlbaEAEnrique De AlbaACAlex Cabello

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates Aperture-X for magnitude estimation in satellite photometry, suggesting improved accuracy.

Key Points

  • This research aims to develop a robust photometry model, Aperture-X, for accurate visual magnitude estimation under challenging conditions.
  • Developed Aperture-X using XGBoost and trained on 1.2 million synthetic patches with varied source morphologies.
  • Integrated multiple physics-derived aperture measurements for improved feature interpretability.
  • Implemented a three-stage curriculum for sim-to-real evaluation involving real star photometry and proxy satellite injections.
  • Aperture-X reduced mean absolute error (MAE) by up to 41% for stars and 48% for satellites under stray light conditions compared to baseline models.
  • Simulated satellite outlier rate dropped to below 0.1% and the characterizable fraction of real stars increased by 12% after adaptation.

Cite This Study

Alba et al. (2026) studied this question.

synapsesocial.com/papers/6a547ff4475c38bf615a52fdhttps://doi.org/10.1016/j.ascom.2026.101165
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Leveraging Synthetic Data for Star and Satellite Photometry2024
  2. 2DeepAP: Deep Learning-based Aperture Photometry Feasibility Assessment and Aperture Size Prediction2025
  3. 3Aperture masking observations in binary stars detection with 1.56-m telescope2024
  4. 4Imaging Simulation for Space Object Detection Using Space-Based Optical Telescopes2026
  5. 5Machine Learning based Pointing Models for Radio/Sub-millimeter Telescopes2024 · 2 citations