Article Details

Abstract

To address the issues of edge blurring and detail loss in existing convolutional neural networks when processing high-density salt-and-pepper noise images, this paper proposes an image denoising algorithm that integrates Fuzzy Logic Deep Learning. The algorithm introduces a learnable soft fuzzy mask module to estimate noise confidence pixel by pixel based on the input image, and combines an improved deep U-Net network to enhance image feature extraction and denoising capabilities. Additionally, to improve the structural fidelity and edge clarity of the restored image, this paper designs a hybrid loss function incorporating Mean Squared Error, Structural Similarity Index Measure (SSIM), and edge loss. Experimental results show that the algorithm exhibits superior performance in both subjective visual effects and objective evaluation metrics.

Keywords

salt-and-pepper noise; image denoising; Fuzzy Logic; Deep Learning; U-Net

References

1. [1] Miao Baoming, Chen Wei, Wu Hang, etc Research on Gaussian and salt and pepper noise removal methods for surgical instrument visual images based on deep learning [J]. Medical and Health Equipment, 2024, 45 (2): 1-7
2. [2] Huang Yufeng, Zhang Xiaobo Image super-resolution method based on deep learning and wavelet transform [J]. Modern Information Technology, 2025, 9 (6): 62-66
3. [3] Wang Pengbo, Dan Wuyang, Li Jun, etc Robust Splicing Forensic Algorithm Against High Intensity Salt and Pepper Noise [J]. Computer Applications, 2024, 44 (10): 3177-3184
4. [4] Liu Guoming Research on Image Salt and Pepper Noise Removal Method Based on Deep Convolutional Neural Network [D]. Anqing: Anqing Normal University, 2022
5. [5] Qiu Junjie Research on Salt and Pepper Noise Filtering Algorithm for Remote Sensing Images [D]. Fuzhou: East China University of Technology, 2020
6. [6] Xu Jian Research on Image Salt and Pepper Noise Removal Algorithm Based on Deep Convolutional Neural Network [D]. Hefei: Hefei University of Technology, 2019
7. [7] Zhao Junpu, Zhang Xiaobo Research on Poisson noise removal algorithm based on multi-scale nonlinear diffusion [J]. Information Technology and Informatization, 2025 (4): 24-30
8. [8] Guixin, Yuan Liuyang, Chi Xiaoni Image Salt and Pepper Noise Filtering Algorithm Based on Adaptive Edge Detection [J]. Practice and Understanding of Mathematics, 2025, 55 (5): 211-225
9. [9] ALI A R，MAHMOUD F. A Deep Convolutional Neural Network for Salt-and-Pepper Noise Removal Using Selective Convolutional blocks [J].Applied Soft Computing，2023，145：110535.
10. [10] SINGH V，DEW R，DHAR，et al. Adaptive Type-2 fuzzy Approach for Filtering Salt and Pepper Noise in Grayscale Images [J].IEEE Transactions on Fuzzy Systems，2018，26（5）：3170-3176.
11. [11] GAO J Q，LI L，REN X D，et al. An Effective Method for Salt and Pepper Noise Removal Based on Algebra and Fuzzy Logic Function [J].Multimedia Tools and Applications，2024，83（4）：9547-9576.
12. [12] SINGH V，AGRAWAL P，SHARMA T，et al. Improved Adaptive Type-2 Fuzzy Filter with Exclusively two Fuzzy Membership Function for Filtering Salt and Pepper Noise [J].Multimedia Tools and Applications，2022，82（13）：20015-20037.
13. [13] ERKAN U，GÖKREM L，ENGINOĞLU S. Different Applied Median Filter in Salt and Pepper Noise [J].Computers & Electrical Engineering，2018，70：789-798.
14. [14] YEH C H，YANG H F，CHEN M J，et al. Image InPainting Based on GAN-Driven Structure-and Texture-aware Learning with Application to Object Removal [J/OL].Applied Soft Computing，2024，161：111748[2025-10-28].https://www.sciencedirect.com/science/article/pii/S1568494624005222.
15. [15] BHATTI U A，YU Z Y，CHANUSSOT J，et al. Local Similarity-Based Spatial Spectral Fusion Hyperspectral Image Classification with Deep CNN and Gabor Filtering [J].IEEE Transactions on Geoscience and Remote Sensing，2021，60：1-15
16. [16] BAKKOURI I，AFDEL K. Correction to Computer-aided Diagnosis(CAD)System Based on Multi-layer Feature Fusion Network for Skin Lesion Recognition in Dermoscopy Images [J].Multimedla Tools and Applications，2020，79（29）：20483-20518
17. [17] FU B，ZHAO X Y，LI Y，et al. A convolutional Neural Networks Denoising Approach for Salt and Pepper Noise [J].Multimedia Tools and Applications，2019，78（21）：30707-30721.
18. [18] GAO J Q，LI L，REN X D，et al. An Effective Method for Salt and Pepper Noise Removal Based on Algebra and Fuzzy Logic Function [J].Multimedia Tools and Applications，2024，83（4）：9547-9576.
19. [19] PUTRA B P E，PRASETYO H，SURYANI E. Residual Transformer Fusion Network for Salt and Pepper Image Denoising [J/OL].arXiv:2502.09000 [cs.CV].[2025-10-28].https://arxiv.org/abs/2502.09000.

About the authors

Tian Huiyuan (2000-), male, Han ethnicity, from Zhoukou, Henan, currently pursuing a master's degree in deep learning; Zhang Xiaobo (1979-), male, Han ethnicity, born in Xianyang, Shaanxi Province, professor, PhD, research interests: machine learning, big data analysis, image processing, natural language processing.