FEUSNet: Fourier Embedded U-Shaped Network for Image Denoising

Li, Xi; Han, Jingwei; Yuan, Quan; Zhang, Yaozong; Fu, Zhongtao; Zou, Miao; Huang, Zhenghua

doi:10.3390/e25101418

Open AccessArticle

FEUSNet: Fourier Embedded U-Shaped Network for Image Denoising

by

Xi Li

^1,2,

Jingwei Han

^1,*,

Quan Yuan

^2,*,

Yaozong Zhang

²,

Zhongtao Fu

²,

Miao Zou

¹ and

Zhenghua Huang

²

¹

School of Electrical and Information Engineering, Wuhan Institute of Technology, Wuhan 430205, China

²

College of Information and Artificial Intelligence, Nanchang Institute of Science and Technology, Nanchang 330108, China

^*

Authors to whom correspondence should be addressed.

Entropy 2023, 25(10), 1418; https://doi.org/10.3390/e25101418

Submission received: 21 July 2023 / Revised: 3 October 2023 / Accepted: 4 October 2023 / Published: 5 October 2023

(This article belongs to the Special Issue Advances in Computer Recognition, Image Processing and Communications, Selected Papers from CORES 2023 and IP&C 2023)

Download

Browse Figures

Review Reports Versions Notes

Abstract

Deep convolution neural networks have proven their powerful ability in comparing many tasks of computer vision due to their strong data learning capacity. In this paper, we propose a novel end-to-end denoising network, termed Fourier embedded U-shaped network (FEUSNet). By analyzing the amplitude spectrum and phase spectrum of Fourier coefficients, we find that low-frequency features of an image are in the former while noise features are in the latter. To make full use of this characteristic, Fourier features are learned and are concatenated as a prior module that is embedded into a U-shaped network to reduce noise while preserving multi-scale fine details. In the experiments, we first present ablation studies on the Fourier coefficients’ learning networks and loss function. Then, we compare the proposed FEUSNet with the state-of-the-art denoising methods in quantization and qualification. The experimental results show that our FEUSNet performs well in noise suppression and preserves multi-scale enjoyable structures, even outperforming advanced denoising approaches.

Keywords: deep convolution neural network; end-to-end denoising network mechanism; Fourier coefficients

Share and Cite

MDPI and ACS Style

Li, X.; Han, J.; Yuan, Q.; Zhang, Y.; Fu, Z.; Zou, M.; Huang, Z. FEUSNet: Fourier Embedded U-Shaped Network for Image Denoising. Entropy 2023, 25, 1418. https://doi.org/10.3390/e25101418

AMA Style

Li X, Han J, Yuan Q, Zhang Y, Fu Z, Zou M, Huang Z. FEUSNet: Fourier Embedded U-Shaped Network for Image Denoising. Entropy. 2023; 25(10):1418. https://doi.org/10.3390/e25101418

Chicago/Turabian Style

Li, Xi, Jingwei Han, Quan Yuan, Yaozong Zhang, Zhongtao Fu, Miao Zou, and Zhenghua Huang. 2023. "FEUSNet: Fourier Embedded U-Shaped Network for Image Denoising" Entropy 25, no. 10: 1418. https://doi.org/10.3390/e25101418

APA Style

Li, X., Han, J., Yuan, Q., Zhang, Y., Fu, Z., Zou, M., & Huang, Z. (2023). FEUSNet: Fourier Embedded U-Shaped Network for Image Denoising. Entropy, 25(10), 1418. https://doi.org/10.3390/e25101418

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

FEUSNet: Fourier Embedded U-Shaped Network for Image Denoising

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI