Volume 18 Issue 5
Sep.  2025
Turn off MathJax
Article Contents
OU Hai-ning, LI Chang-di, ZENG Rui-bin, WU Yan-feng, LIU Jia-ning, CHENG Peng. DnCNN-RM: an adaptive SAR image denoising algorithm based on residual networks[J]. Chinese Optics, 2025, 18(5): 1209-1218. doi: 10.37188/CO.EN-2024-0028
Citation: OU Hai-ning, LI Chang-di, ZENG Rui-bin, WU Yan-feng, LIU Jia-ning, CHENG Peng. DnCNN-RM: an adaptive SAR image denoising algorithm based on residual networks[J]. Chinese Optics, 2025, 18(5): 1209-1218. doi: 10.37188/CO.EN-2024-0028

DnCNN-RM: an adaptive SAR image denoising algorithm based on residual networks

cstr: 32171.14.CO.EN-2024-0028
Funds:  Supported by Open Fund of the Anhui Provincial Key Laboratory of Machine Vision Inspection and Perception (No. KLMV1-2024-HIT-01)
More Information
  • Author Bio:

    OU Hai-ning (1979—), male, born in Putian, Fujian Province, holds a Bachelor's degree. He received his Bachelor's degree from Huaqiao University in 2002. He is employed at Meizhouwan Vocational Technology College. His primarily engaged in research related to computer vision. E-mail: mzyxqb@126.com

    LI Chang-di (1995—), male, born in Huaibei, Anhui Province, holds a Master's degree. He received his Master's degree from Soochow University in 2020. He is employed at Suzhou Duty Ratio Information Technology Co., Ltd. and is primarily engaged in research related to artificial intelligence. E-mail: lcd33480275@163.com

    ZENG Rui-bin (1994—), male, born in Ningde, Fujian Province, holds a Master's degree. He received his Master's degree from the University of Chinese Academy of Sciences in 2023. He is employed at Meizhouwan Vocational Technology College. His primarily engaged in research related to artificial intelligence. E-mail: zengruibin19@mails.ucas.ac.cn

  • Corresponding author: lcd33480275@163.comzengruibin19@mails.ucas.ac.cn
  • Received Date: 10 Oct 2024
  • Rev Recd Date: 15 Nov 2024
  • Accepted Date: 17 Jan 2025
  • Available Online: 26 Jan 2025
  • In the field of image processing, the analysis of Synthetic Aperture Radar (SAR) images is crucial due to its broad range of applications. However, SAR images are often affected by coherent speckle noise, which significantly degrades image quality. Traditional denoising methods, typically based on filter techniques, often face challenges related to inefficiency and limited adaptability. To address these limitations, this study proposes a novel SAR image denoising algorithm based on an enhanced residual network architecture, with the objective of enhancing the utility of SAR imagery in complex electromagnetic environments. The proposed algorithm integrates residual network modules, which directly process the noisy input images to generate denoised outputs. This approach not only reduces computational complexity but also mitigates the difficulties associated with model training. By combining the Transformer module with the residual block, the algorithm enhances the network's ability to extract global features, offering superior feature extraction capabilities compared to CNN-based residual modules. Additionally, the algorithm employs the adaptive activation function Meta-ACON, which dynamically adjusts the activation patterns of neurons, thereby improving the network's feature extraction efficiency. The effectiveness of the proposed denoising method is empirically validated using real SAR images from the RSOD dataset. The proposed algorithm exhibits remarkable performance in terms of EPI, SSIM, and ENL, while achieving a substantial enhancement in PSNR when compared to traditional and deep learning-based algorithms. The PSNR performance is enhanced by over twofold. Moreover, the evaluation of the MSTAR SAR dataset substantiates the algorithm's robustness and applicability in SAR denoising tasks, with a PSNR of 25.2021 being attained. These findings underscore the efficacy of the proposed algorithm in mitigating speckle noise while preserving critical features in SAR imagery, thereby enhancing its quality and usability in practical scenarios.

     

  • loading
  • [1]
    THEODOSIOU A, LÓPEZ-DEKKER P. On the sensitivity to height and motion of bistatic SAR interferometry: a spectral view[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62: 5212212.
    [2]
    WU Y M, SUO Y X, MENG Q B, et al. FAIR-CSAR: a benchmark dataset for fine-grained object detection and recognition based on single-look complex SAR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2025, 63: 5201022.
    [3]
    ZHAO CH H, LI K Y, WANG L, et al. A rapid SAR image simulation method for ship wakes coupled with sea waves using fluid velocity potential[J]. IEEE Signal Processing Letters, 2025, 32: 271-275. doi: 10.1109/LSP.2024.3514804
    [4]
    GOU SH P, LI D B, HAI D, et al. Spectral clustering with eigenvalue similarity metric method for POL-SAR image segmentation of land cover[J]. Journal of Geographic Information System, 2018, 10: 150-164. doi: 10.4236/jgis.2018.101007
    [5]
    PETTINATO S, SANTI E, PALOSCIA S, et al. Large area robust identification of snow cover from multitemporal COSMO-SkyMed images[J]. Proceedings of SPIE, 2015, 9642: 964204.
    [6]
    LI CH D, LI G Y, SONG Y CH, et al. Fast forest fire detection and segmentation application for UAV-assisted mobile edge computing system[J]. IEEE Internet of Things Journal, 2024, 11(16): 26690-26699. doi: 10.1109/JIOT.2023.3311950
    [7]
    QIN CH, ZHANG L P, WANG X Q, et al. RDB-DINO: an improved end-to-end transformer with refined de-noising and boxes for small-scale ship detection in SAR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2025, 63: 5200517.
    [8]
    JOSHI G P, ALENEZI F, THIRUMOORTHY G, et al. Ensemble of deep learning-based multimodal remote sensing image classification model on unmanned aerial vehicle networks[J]. Mathematics, 2021, 9(22): 2984. doi: 10.3390/math9222984
    [9]
    SUN Y X, FENG SH SH, YE Y M, et al. Multisensor fusion and explicit semantic preserving-based deep hashing for cross-modal remote sensing image retrieval[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 60: 5219614.
    [10]
    THAKUR R S, YADAV R N, GUPTA L. State-of-art analysis of image denoising methods using convolutional neural networks[J]. IET Image Processing, 2019, 13(13): 2367-2380. doi: 10.1049/iet-ipr.2019.0157
    [11]
    DUAN Y P, TAO X M, HAN C Y, et al. Multi-scale convolutional neural network for SAR image semantic segmentation[C]. 2018 IEEE Global Communications Conference (GLOBECOM), IEEE, 2018: 1-6.
    [12]
    WANG CH, YIN ZH X, MA X SH, et al. SAR image despeckling based on block-matching and noise-referenced deep learning method[J]. Remote Sensing, 2022, 14(4): 931. doi: 10.3390/rs14040931
    [13]
    GUO Q, LIU J N, KALIUZHNYI M. YOLOX-SAR: high-precision object detection system based on visible and infrared sensors for SAR remote sensing[J]. IEEE Sensors Journal, 2022, 22(17): 17243-17253. doi: 10.1109/JSEN.2022.3186889
    [14]
    WANG P Y, ZHANG H, PATEL V M. SAR image despeckling using a convolutional neural network[J]. IEEE Signal Processing Letters, 2017, 24(12): 1763-1767. doi: 10.1109/LSP.2017.2758203
    [15]
    ZHANG Q, YUAN Q Q, LI J, et al. Learning a dilated residual network for SAR image despeckling[J]. Remote Sensing, 2018, 10(2): 196. doi: 10.3390/rs10020196
    [16]
    王曦. 基于深度学习的SAR图像目标识别方法研究[D]. 成都: 电子科技大学, 2022.

    WANG X. Research on SAR image target recognition method based on deep learning[D]. Chengdu: University of Electronic Science and Technology of China, 2022. (in Chinese).
  • 加载中

Catalog

    通讯作者: 陈斌, bchen63@163.com
    • 1. 

      沈阳化工大学材料科学与工程学院 沈阳 110142

    1. 本站搜索
    2. 百度学术搜索
    3. 万方数据库搜索
    4. CNKI搜索

    Figures(6)  / Tables(4)

    Article views(225) PDF downloads(22) Cited by()
    Proportional views
    Related

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return