Noise2Kernel: Adaptive Self-Supervised Blind Denoising using a Dilated Convolutional Kernel Architecture

TL;DR

Noise2Kernel introduces an adaptive self-supervised blind denoising method using dilated convolutional kernels, effectively handling extreme and hybrid noise without prior noise statistics, and overcoming brightness shifting issues.

Contribution

It presents a novel dilated convolutional architecture with invariant properties and an adaptive self-supervision loss for improved blind denoising without random masking.

Findings

Outperforms state-of-the-art denoising methods on various noise types.

Effectively removes salt-and-pepper and hybrid noise without prior noise knowledge.

Addresses brightness shifting issues in extreme noise scenarios.

Abstract

With the advent of recent advances in unsupervised learning, efficient training of a deep network for image denoising without pairs of noisy and clean images has become feasible. However, most current unsupervised denoising methods are built on the assumption of zero-mean noise under the signal-independent condition. This assumption causes blind denoising techniques to suffer brightness shifting problems on images that are greatly corrupted by extreme noise such as salt-and-pepper noise. Moreover, most blind denoising methods require a random masking scheme for training to ensure the invariance of the denoising process. In this paper, we propose a dilated convolutional network that satisfies an invariant property, allowing efficient kernel-based training without random masking. We also propose an adaptive self-supervision loss to circumvent the requirement of zero-mean constraint, which is specifically effective in removing salt-and-pepper or hybrid noise where a prior knowledge of noise statistics is not readily available. We demonstrate the efficacy of the proposed method by comparing it with state-of-the-art denoising methods using various examples.