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

TL;DR

The paper introduces ASTERIS, a self-supervised transformer-based denoising method that enhances astronomical imaging detection limits by leveraging spatiotemporal data, leading to improved detection of faint features.

Contribution

ASTERIS is a novel self-supervised transformer model that integrates across multiple exposures to improve detection limits in astronomical images.

Findings

ASTERIS improves detection limits by 1.0 magnitude at 90% completeness.

It preserves point spread function and photometric accuracy.

Identifies three times more high-redshift galaxy candidates in JWST images.

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.