Regularizing Instabilities in Image Reconstruction Arising from Learned Denoisers

TL;DR

This paper addresses the instability issues in learned image reconstruction algorithms caused by adversarial noise, proposing regularization techniques to improve their stability and reliability.

Contribution

The authors extend classical regularization concepts to learned reconstruction algorithms, introducing methods to stabilize them against adversarial noise.

Findings

Regularization techniques improve stability of learned reconstructions.

Enhanced robustness against adversarial noise in image reconstruction.

Applicable to popular learned reconstruction algorithms.

Abstract

It's well-known that inverse problems are ill-posed and to solve them meaningfully one has to employ regularization methods. Traditionally, popular regularization methods have been the penalized Variational approaches. In recent years, the classical regularized-reconstruction approaches have been outclassed by the (deep-learning-based) learned reconstruction algorithms. However, unlike the traditional regularization methods, the theoretical underpinnings, such as stability and regularization, have been insufficient for such learned reconstruction algorithms. Hence, the results obtained from such algorithms, though empirically outstanding, can't always be completely trusted, as they may contain certain instabilities or (hallucinated) features arising from the learned process. In fact, it has been shown that such learning algorithms are very susceptible to small (adversarial) noises in the data and can lead to severe instabilities in the recovered solution, which can be quite different than the inherent instabilities of the ill-posed (inverse) problem. Whereas, the classical regularization methods can handle such (adversarial) noises very well and can produce stable recovery. Here, we try to present certain regularization methods to stabilize such (unstable) learned reconstruction methods and recover a regularized solution, even in the presence of adversarial noises. For this, we need to extend the classical notion of regularization and incorporate it in the learned reconstruction algorithms. We also present some regularization techniques to regularize two of the most popular learning reconstruction algorithms, the Learned Post-Processing Reconstruction and the Learned Unrolling Reconstruction.