Self-supervised denoising for massive noisy images

TL;DR

This paper introduces a deep learning model for denoising massive noisy images that does not require clean samples or noise calibration, effective across diverse real-world applications.

Contribution

It presents a novel self-supervised denoising approach that operates without prior noise models or clean training data, applicable to various imaging domains.

Findings

Effective on atomic and astronomy images

No need for noise calibration or clean samples

Performs well across different real-world noisy images

Abstract

We propose an effective deep learning model for signal reconstruction, which requires no signal prior, no noise model calibration, and no clean samples. This model only assumes that the noise is independent of the measurement and that the true signals share the same structured information. We demonstrate its performance on a variety of real-world applications, from sub-\r{A}ngstr\"{o}m resolution atomic images to sub-arcsecond resolution astronomy images.