Deeper detection limits in astronomical imaging using self-supervised spatiotemporal denoising

2026-02-19

AI Summary

This work addresses the fundamental limitation in astronomical imaging imposed by spatiotemporally correlated noise, which hampers the detection of faint celestial sources. We propose ASTERIS—the first self-supervised Transformer-based denoising method tailored for astronomical imaging—that effectively suppresses such correlated noise by leveraging spatiotemporal information across multiple exposures, without requiring labeled training data. While preserving the point spread function and photometric accuracy, ASTERIS significantly enhances detection depth: in simulated data, it improves the detection limit by 1.0 magnitude at 90% completeness and purity, and in real observations from JWST and Subaru, it reveals additional low-surface-brightness structures, gravitational lensing arcs, and triples the number of candidate galaxies at redshifts greater than 9.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Unsupervised & Self-Supervised Learning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

The detection limit of astronomical imaging observations is limited by several noise sources. Some of that noise is correlated between neighbouring image pixels and exposures, so in principle could be learned and corrected. We present an astronomical self-supervised transformer-based denoising algorithm (ASTERIS), that integrates spatiotemporal information across multiple exposures. Benchmarking on mock data indicates that ASTERIS improves detection limits by 1.0 magnitude at 90% completeness and purity, while preserving the point spread function and photometric accuracy. Observational validation using data from the James Webb Space Telescope (JWST) and Subaru telescope identifies previously undetectable features, including low-surface-brightness galaxy structures and gravitationally-lensed arcs. Applied to deep JWST images, ASTERIS identifies three times more redshift > 9 galaxy candidates, with rest-frame ultraviolet luminosity 1.0 magnitude fainter, than previous methods.

Problem

• astronomical imaging
• detection limit
• noise
• low-surface-brightness structures
• high-redshift galaxies

Innovation

• self-supervised denoising
• spatiotemporal transformer
• astronomical imaging
• detection limit enhancement
• JWST data analysis