SynFSNet: Synergizing Frequency Domain and Spatial Domain Modeling Network for Image Deraining

doi:10.19678/j.issn.1000-3428.0253484

Computer Engineering ›› 2026, Vol. 52 ›› Issue (8): 58-70. doi: 10.19678/j.issn.1000-3428.0253484

Special Issue: ；

• AI-enabled Industrial Innovation in Building Materials, Construction, Automotive and Transportation • Previous Articles Next Articles

SynFSNet: Synergizing Frequency Domain and Spatial Domain Modeling Network for Image Deraining

LI Yamin¹^,²^,³, XIANG Wen¹^,²^,³, LIU Yuting¹^,²^,³, XIANG Yao¹^,⁴^,*()

1. School of Computer Science, Hubei University, Wuhan 430062, Hubei, China
2. Key Laboratory of Intelligent Sensing System and Security, Ministry of Education, Wuhan 430062, Hubei, China
3. Hubei Key Laboratory of Big Data Intelligent Analysis and Application, Hubei University, Wuhan 430062, Hubei, China
4. School of Computer Science and Artificial Intelligence, Wuhan University of Technology, Wuhan 430070, Hubei, China

Received:2025-12-25 Revised:2026-05-16 Online:2026-08-15 Published:2026-07-10
Contact: XIANG Yao

SynFSNet: 频域与空间域协同建模的图像去雨网络

李亚敏¹^,²^,³, 向稳¹^,²^,³, 刘钰婷¹^,²^,³, 向尧¹^,⁴^,*()

1. 湖北大学计算机学院, 湖北武汉 430062
2. 智能感知系统与安全教育部重点实验室, 湖北武汉 430062
3. 大数据智能分析与行业应用湖北省重点实验室(湖北大学), 湖北武汉 430062
4. 武汉理工大学计算机与人工智能学院, 湖北武汉 430070

通讯作者: 向尧
作者简介:
李亚敏(CCF专业会员)，女，副教授、博士，主研方向为计算机视觉、图像处理
向稳，硕士研究生
刘钰婷，硕士研究生
向尧(通信作者)，讲师、博士
基金资助:
湖北省自然科学基金面上项目(JCZRYB202501223)

Abstract

Abstract:

Image deraining aims to restore high-quality clean images from rain-degraded inputs and is a key technique for improving the robustness of outdoor vision systems. Existing Convolutional Neural Network (CNN)-based methods are limited by local receptive fields and thus struggle to effectively model long-range dependencies. Although some Transformer-based methods enhance global modeling capability, they show limited ability in handling multi-scale and directional rain streaks as well as recovering local high-frequency details. Moreover, existing frequency-domain enhancement methods often treat the frequency domain merely as an auxiliary representation, which lack fine-grained collaborative modeling of the complementary relationship between the spatial and frequency domains. To address these issues, this study proposes a Synergizing Frequency and Spatial Network (SynFSNet) for image deraining. Unlike existing methods that primarily use the frequency domain as a global enhancement tool, the proposed method uses the directional, density-related, and scale-varying characteristics of rain streaks in the frequency domain. Specifically, a Multi-Scale Fourier Fusion Module (MSFFM) is designed to progressively suppress rain streaks in a coarse-to-fine manner. Additionally, a Fourier Shaped Attention Module (FSAM) is introduced in which Rectangular Filter Attention (RFA) and Square Filter Attention (SFA) are used to enhance direction-sensitive and local-structure-sensitive frequency-domain representations, respectively. Furthermore, a dual-domain loss is employed to jointly constrain spatial structural restoration and frequency-domain consistency, thereby improving restoration quality in complex scenes. Experimental results show that SynFSNet achieves superior performance on multiple synthetic and real-world datasets, verifying the effectiveness of collaborative spatial—frequency domain modeling for complex image deraining tasks.

Key words: image deraining, frequency-domain feature, multi-scale Fourier fusion, filter attention, Fourier transform

摘要：

图像去雨旨在从受雨迹干扰的退化图像中恢复高质量的无雨图像, 是提升户外视觉系统鲁棒性的关键技术。现有基于卷积神经网络(CNN)的方法受限于局部感受野, 难以有效建模长程依赖关系; 部分基于Transformer的方法虽增强了全局建模能力, 但对复杂雨迹的多尺度、方向性及局部高频细节恢复仍存在不足; 已有频域增强方法也多将频域作为辅助表示手段, 缺乏针对空间域与频域互补关系的细粒度协同建模。为此, 本文提出了一种频域与空间域协同建模的图像去雨网络(SynFSNet)。不同于现有方法主要将频域作为全局增强工具, 本文从雨迹在频域中的方向性、密度性和尺度变化出发, 通过多尺度傅里叶融合模块(MSFFM)实现由粗到细的渐进式雨迹抑制, 并通过傅里叶形状注意模块(FSAM)中的矩形滤波器注意力(RFA)和方形滤波器注意力(SFA)分别增强方向敏感与局部结构敏感的频域表示。进一步地, 网络通过双域损失函数同时约束空间域结构与频域一致性, 以提升复杂场景下的恢复质量。实验结果表明, SynFSNet在多个合成数据集和真实场景数据集上均取得了较优性能, 验证了频域与空间域协同建模在复杂图像去雨任务中的有效性。

关键词: 图像去雨, 频域特征, 多尺度傅里叶融合, 滤波器注意力, 傅里叶变换

LI Yamin, XIANG Wen, LIU Yuting, XIANG Yao. SynFSNet: Synergizing Frequency Domain and Spatial Domain Modeling Network for Image Deraining[J]. Computer Engineering, 2026, 52(8): 58-70.

李亚敏, 向稳, 刘钰婷, 向尧. SynFSNet: 频域与空间域协同建模的图像去雨网络[J]. 计算机工程, 2026, 52(8): 58-70.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0253484

https://www.ecice06.com/EN/Y2026/V52/I8/58

Figures/Tables 17

Fig.1 Overall framework of SynFSNet

Fig.2 Rectangle filter attention module

Fig.3 Square filter attention

Fig.4 Analysis of frequency-domain perspective and rain streak degradation phenomena

Fig.5 Comparison of PSNR and parameters

Fig.6 Visual comparisons of derained results on the Rain200L dataset

Fig.7 Visual comparisons of derained results on the Rain200H dataset

Fig.8 Visual comparisons of derained results on the real-world dataset SPA-Data

Fig.9 Qualitative results in real-world low-light and rainy-snowy scenes

References 48

1	WANG K G, WANG T M, QU J C, et al. An end-to-end cascaded image deraining and object detection neural network. IEEE Robotics and Automation Letters, 2022, 7(4): 9541- 9548. doi: 10.1109/LRA.2022.3192200
2	白邵宙, 张浩, 赵景波, 等. 基于颜色均衡与特征融合的水下图像增强框架. 计算机工程, 2025, 51(10): 336- 345. doi: 10.19678/j.issn.1000-3428.0069600
	BAI S Z, ZHANG H, ZHAO J B, et al. Underwater image enhancement frame based on color balance and feature fusion. Computer Engineering, 2025, 51(10): 336- 345. doi: 10.19678/j.issn.1000-3428.0069600
3	JIANG Y D, ZHU B, ZHAO X M, et al. Pixel-wise content attention learning for single-image deraining of autonomous vehicles. Expert Systems with Applications, 2023, 224, 119990. doi: 10.1016/j.eswa.2023.119990
4	ALI A M, BENJDIRA B, KOUBAA A, et al. Vision transformers in image restoration: a survey. Sensors, 2023, 23(5): 2385- 2385. doi: 10.3390/s23052385
5	LI J H, HU J C, FU P Y, et al. Ultra-fast deraining plugin for vision-based perception of autonomous driving. IEEE Transactions on Intelligent Transportation Systems, 2025, 26(1): 1227- 1240. doi: 10.1109/TITS.2024.3503556
6	LIANG Z, RUAN R, WANG C J, et al. Single image quality improvement via joint local structure dehazing and local texture enhancement. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62, 4210117. doi: 10.1109/TGRS.2024.3451471
7	TAO W H, SHAO F F, SU L J, et al. An analytical model for simulating the rainfall-interception-infiltration-runoff process with non-uniform rainfall. Journal of Environmental Management, 2023, 344, 118490. doi: 10.1016/j.jenvman.2023.118490
8	LIU R W, LU Y X, GUO Y, et al. AiOENet: all-in-one low-visibility enhancement to improve visual perception for intelligent marine vehicles under severe weather conditions. IEEE Transactions on Intelligent Vehicles, 2024, 9(2): 3811- 3826. doi: 10.1109/TIV.2023.3347952
9	SU Z P, ZHANG Y X, SHI J H, et al. A survey of single image rain removal based on deep learning. ACM Computing Surveys, 2024, 56(4): 1- 35. doi: 10.1145/3625818
10	TAO W Y, YAN X F, WANG Y Z, et al. MFFDNet: single image deraining via dual-channel mixed feature fusion. IEEE Transactions on Instrumentation and Measurement, 2024, 73, 5005813.
11	CHENG D, LI Y, ZHANG D W, et al. Progressive negative enhancing contrastive learning for image dehazing and beyond. IEEE Transactions on Multimedia, 2024, 26, 8783- 8798. doi: 10.1109/TMM.2024.3382493
12	杨士钺. 基于卷积神经网络的图像去雨方法研究及应用[D]. 大庆: 东北石油大学, 2024.
	YANG S Y. Research and application of image rain removal method based on convolutional neural network[D]. Daqing: Northeast Petroleum University, 2024. (in Chinese)
13	WANG Z D, CUN X D, BAO J M, et al. Uformer: a general U-shaped transformer for image restoration[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New Orleans, USA: IEEE Press, 2022: 17662-17672.
14	朱凯, 李理, 张彤, 等. 基于Transformer的多阶段运动模糊图像修复网络. 计算机工程, 2024, 50(9): 276- 285. doi: 10.19678/j.issn.1000-3428.0068246
	ZHU K, LI L, ZHANG T, et al. Multi-stage motion blur image restoration network based on Transformer. Computer Engineering, 2024, 50(9): 276- 285. doi: 10.19678/j.issn.1000-3428.0068246
15	YANG J K, WANG J Y, LI Y B, et al. Image deraining algorithm based on multi-scale features. Applied Sciences, 2024, 14(13): 5548. doi: 10.3390/app14135548
16	GAO N, JIANG X Y, ZHANG X H, et al. Efficient frequency-domain image deraining with contrastive regularization[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2025: 240-257.
17	杨佳璇. 基于空频双域特征提取的图像去雨关键技术研究[D]. 天津: 天津工业大学, 2025.
	YANG J X. Research on key technologies of image rain removal based on spatial-frequency dual-domain feature extraction[D]. Tianjin: Tianjin Polytechnic University, 2025. (in Chinese)
18	LYU P F, YU X S, YEUNG P H, et al. Deep Fourier-embedded network for RGB and thermal salient object detection[EB/OL]. [2025-11-02]. https://arxiv.org/abs/2411.18409.
19	DING X H, CHEN L Q, ZHENG X H, et al. Single image rain and snow removal via guided L0 smoothing filter. Multimedia Tools and Applications, 2016, 75(5): 2697- 2712. doi: 10.1007/s11042-015-2657-7
20	ZHANG H, PATEL V M. Convolutional sparse and low-rank coding-based rain streak removal[C]//Proceedings of the IEEE Winter Conference on Applications of Computer Vision. Santa Rosa, USA: IEEE Press, 2017: 1259-1267.
21	JIANG K, WANG Z Y, YI P, et al. Multi-scale progressive fusion network for single image deraining[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, USA: IEEE Press, 2020: 8343-8352.
22	WANG H, XIE Q, ZHAO Q, et al. RCDNet: an interpretable rain convolutional dictionary network for single image deraining. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(6): 8668- 8682. doi: 10.1109/TNNLS.2022.3231453
23	CHEN X, LI H, LI M Q, et al. Learning a sparse transformer network for effective image deraining[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Vancouver, Canada: IEEE Press, 2023: 5896-5905.
24	XIAO J, FU X Y, LIU A P, et al. Image de-raining transformer. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(11): 12978- 12995. doi: 10.1109/TPAMI.2022.3183612
25	ZHANG T S, LIU P P, ZHAO M, et al. DMFourLLIE: dual-stage and multi-branch Fourier network for low-light image enhancement[C]//Proceedings of the 32nd ACM International Conference on Multimedia. New York, USA: ACM Press, 2024: 7434-7443.
26	ZHOU M, HUANG J, GUO C L, et al. Fourmer: an efficient global modeling paradigm for image restoration[C]//Proceedings of the International Conference on Machine Learning. [S. l.]: PMLR, 2023: 42589-42601.
27	HUANG J, LIU Y J, ZHAO F, et al. Deep Fourier-based exposure correction network with spatial-frequency interaction[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2022: 163-180.
28	YU H, ZHENG N S, ZHOU M, et al. Frequency and spatial dual guidance for image dehazing[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2022: 181-198.
29	JIANG X Y, ZHANG X H, GAO N, et al. When fast Fourier transform meets transformer for image restoration[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2025: 381-402.
30	LI S Y, CAI S Q, TENG S, et al. Weather-robust spatial-frequency decoupling transformer for crack segmentation. IEEE Sensors Journal, 2025, 25(21): 39881- 39892. doi: 10.1109/JSEN.2025.3554650
31	PARK N, KIM S. How do vision transformers work?[EB/OL]. [2025-11-02]. https://arxiv.org/abs/2202.06709.
32	CUI Y N, REN W Q, CAO X C, et al. Revitalizing convolutional network for image restoration. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(12): 9423- 9438. doi: 10.1109/TPAMI.2024.3419007
33	LOSHCHILOV I, HUTTER F. Decoupled weight decay regularization[EB/OL]. [2025-11-02]. https://arxiv.org/abs/1711.05101.
34	LOSHCHILOV I, HUTTER F. SGDR: stochastic gradient descent with warm restarts[EB/OL]. [2025-11-02]. https://arxiv.org/pdf/1608.03983.
35	HUYNH-THU Q, GHANBARI M. Scope of validity of PSNR in image/video quality assessment. Electronics Letters, 2008, 44(13): 800- 801. doi: 10.1049/el:20080522
36	WANG Z, BOVIK A C, SHEIKH H R, et al. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing, 2004, 13(4): 600- 612. doi: 10.1109/TIP.2003.819861
37	YANG W H, TAN R T, FENG J S, et al. Deep joint rain detection and removal from a single image[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE Press, 2017: 1685-1694.
38	ZHANG H, PATEL V M. Density-aware single image de-raining using a multi-stream dense network[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: IEEE Press, 2018: 695-704.
39	FU X Y, HUANG J B, ZENG D L, et al. Removing rain from single images via a deep detail network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE Press, 2017: 1715-1723.
40	WANG T Y, YANG X, XU K, et al. Spatial attentive single-image deraining with a high quality real rain dataset[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, USA: IEEE Press, 2020: 12262-12271.
41	LI X, WU J L, LIN Z C, et al. Recurrent squeeze-and-excitation context aggregation net for single image deraining[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2018: 262-277.
42	REN D W, ZUO W M, HU Q H, et al. Progressive image deraining networks: a better and simpler baseline[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, USA: IEEE Press, 2019: 3932-3941.
43	ZAMIR S W, ARORA A, KHAN S, et al. Multi-stage progressive image restoration[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville, USA: IEEE Press, 2021: 14821-14831.
44	YI Q S, LI J C, DAI Q Y, et al. Structure-preserving deraining with residue channel prior guidance[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Montreal, Canada: IEEE Press, 2021: 4218-4227.
45	ZAMIR S W, ARORA A, KHAN S, et al. Restormer: efficient transformer for high-resolution image restoration[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans, USA: IEEE Press, 2022: 5718-5729.
46	CHEN X, PAN J S, LU J Y, et al. Hybrid CNN-transformer feature fusion for single image deraining. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37(1): 378- 386. doi: 10.1609/aaai.v37i1.25111
47	ZHU Z L, ZENG T, YANG T, et al. DeRainMamba: a frequency-aware state space model with detail enhancement for image deraining. IEEE Signal Processing Letters, 2025, 32, 3869- 3873. doi: 10.1109/LSP.2025.3616020
48	JIANG X Y, GAO N, DOU H K, et al. Global modeling matters: a fast, lightweight, and effective baseline for efficient image restoration. IEEE Transactions on Image Processing, 2026, 35, 2740- 2754. doi: 10.1109/TIP.2026.3671691

[1]	HUO Jiuyuan, LI Xin, CHANG Chen, ZHANG Yaonan. Dual-Channel Rolling Bearing Fault Diagnosis Method Based on ACNN-LFSwin Transformer [J]. Computer Engineering, 2026, 52(5): 430-444.
[2]	WANG Zihan, WANG Dan. A Partitioned Space Projection Particle Swarm Optimization Algorithm for Multi-Satellite Planning [J]. Computer Engineering, 2026, 52(3): 403-419.
[3]	ZHANG Xingpeng, HE Dong, YANG Mo, YE Hangbin. Nucleus Segmentation Based on Multiscale Attention and Data Augmentation [J]. Computer Engineering, 2025, 51(2): 387-396.
[4]	DENG Xiangyu, Lü Yahui, CHEN Yan. Frequency Domain Characteristics Analysis of Non-Coupled PCNN [J]. Computer Engineering, 2022, 48(6): 213-221.
[5]	WANG Fei, XU Zhan, ZHI Ruxin, CHEN Jinhui. OFDM-UWB Channel Environment Recognition Based on Convolutional Neural Network [J]. Computer Engineering, 2021, 47(7): 161-167.
[6]	DU Xinjun, LIU Pengfei. Research on Joint Optimization Method of Control Parameters in WFRFT Cognitive Communication System [J]. Computer Engineering, 2021, 47(12): 177-184.
[7]	DANG Xiaochao, ZHANG Jinlong, HAO Zhanjun, AN Ying. Non-Contact Respiration Detection Method Based on IR-UWB Radar [J]. Computer Engineering, 2021, 47(11): 175-184.
[8]	ZHAO Hui, WEI Weibo, PAN Zhenkuan, JI Lianshun. Research on Image Dehazing Based on Dark Channel Prior and Variational Regularization [J]. Computer Engineering, 2021, 47(10): 214-220.
[9]	YUAN Wenhao, SHI Yunlong, HU Shaodong, LOU Yingxi. A Speech Enhancement Approach Based on Fusion of Time-Domain and Frequency-Domain Features [J]. Computer Engineering, 2021, 47(10): 75-81.
[10]	GOU Zhihang, LIU Jianfeng, HU Jinlong, FENG Xuelin, WANG Zongwei. Cyclostationary Detection Method Based on Single Section [J]. Computer Engineering, 2021, 47(1): 196-202.
[11]	LI Mengdong,SHAO Yufang,SUN Yuqing,LI Jie. Efficiency Analysis of SWIFFT Algorithm [J]. Computer Engineering, 2019, 45(1): 109-114.
[12]	MEI Yuan,ZHANG Xiamiao. Fingerprint Enhancement Algorithm Based on Surface Transformation Under Fourier Frequency Domain [J]. Computer Engineering, 2018, 44(8): 251-256.
[13]	FENG Haoran,RUAN Huailin. 2-D DOA Estimation of Wide-band Linear Frequency Modulation Signals with Spatial Under-sampling [J]. Computer Engineering, 2018, 44(8): 296-300.
[14]	CHENG Peng,LIU Aijun,LIANG Xiaohu,WANG Ke,CAI Biao. FTN Signal Carrier Synchronization Based on DFT and Quadratic Polynomial Fitting [J]. Computer Engineering, 2018, 44(6): 86-92.
[15]	LIU Zhong,LI Lichun,LI Huiqi. A Sparsity Adaptive Sparse Fourier Transform Algorithm [J]. Computer Engineering, 2018, 44(2): 141-146.

Please choose a citation manager

Content to export