KRNET: Image Denoising with Kernel Regulation Network

TL;DR

KRNET introduces a novel deep CNN architecture with a kernel regulation module that effectively combines different kernel sizes, significantly improving multi-channel and realistic image denoising performance over existing methods.

Contribution

The paper proposes KR-block, a kernel regulation module that enhances CNN regularization by integrating large and small kernels, leading to a new deep denoising network KRNET.

Findings

KRNET outperforms state-of-the-art methods on various noise types.

KRNET achieves significant performance gains across multiple noise levels.

The kernel regulation module effectively estimates features for denoising.

Abstract

One popular strategy for image denoising is to design a generalized regularization term that is capable of exploring the implicit prior underlying data observation. Convolutional neural networks (CNN) have shown the powerful capability to learn image prior information through a stack of layers defined by a combination of kernels (filters) on the input. However, existing CNN-based methods mainly focus on synthetic gray-scale images. These methods still exhibit low performance when tackling multi-channel color image denoising. In this paper, we optimize CNN regularization capability by developing a kernel regulation module. In particular, we propose a kernel regulation network-block, referred to as KR-block, by integrating the merits of both large and small kernels, that can effectively estimate features in solving image denoising. We build a deep CNN-based denoiser, referred to as KRNET, via concatenating multiple KR-blocks. We evaluate KRNET on additive white Gaussian noise (AWGN), multi-channel (MC) noise, and realistic noise, where KRNET obtains significant performance gains over state-of-the-art methods across a wide spectrum of noise levels.