结合扩张卷积的残差网络SAR图像去噪Residual network combined with dilated convolution for SAR image denoising
申兴成,杨学志,董张玉,陈鲸
摘要(Abstract)：
针对现今的合成孔径雷达图像去噪算法对图像去噪后结构保持不足的问题,该文提出了一种增强的残差卷积神经网络(ERCNN)。首先,ERCNN构建了增强的残差卷积模块,该模块通过结合卷积和扩张卷积,扩大网络的感受野;其次,通过加入残差连接将局部特征信息从低层传递到更高层,减少梯度消失问题;再次,使用批归一化加速模型的训练速度;最后通过残差学习的策略,形成端到端的映射以去除噪声。通过仿真和真实合成孔径雷达图像进行实验的结果表明,所提出的ERCNN具有优于现有方法的性能,在去噪的同时还可以保留更多的细节信息,且拥有高效的计算效率。
关键词(KeyWords)： SAR图像;去噪;扩张卷积;残差连接;结构保持;残差学习
基金项目(Foundation): 国家自然科学基金项目(41601452);; 安徽省科技攻关计划项目(202004a07020030)
作者(Author): 申兴成,杨学志,董张玉,陈鲸
DOI: 10.16251/j.cnki.1009-2307.2021.12.015
参考文献(References)：
- [1] LEE J S.Digital image enhancement and noise filtering by using local statistics[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1980,2(2):165-168.
- [2] SOLBO S,ELTOFT T.Homomorphic wavelet-based statistical despeckling of SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(4):711-721.
- [3] 张倩.基于双重离散小波变换的遥感图像去噪算法[J].国土资源遥感，2015,27(4):14-20.(ZHANG Qian.Remote sensing image de-noising algorithm based on double discrete wavelet transform[J].Remote Sensing for Land and Resources,2015,27(4):14-20.)
- [4] 殷明，刘卫，吴江敏.NSCT域统计模型的SAR图像相干斑抑制[J].测绘科学，2013,38(5):77-79.(YING Ming,LIU Wei,WU Jiangmin.SAR image despeckling based on statistical model in NSCT domain[J].Science of Surveying and Mapping,2013,38(5):77-79.)
- [5] BUADES A,COLL B,MOREL J M.A non-local algorithm for image denoising[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR’05).San Diego,CA,USA:IEEE,2005:60-65.
- [6] DELEDALLE C A,DENIS L,TUPIN F.Iterative weighted maximum likelihood denoising with probabilistic patch-based weights[J].IEEE Transactions on Image Processing,2009,18(12):2661-2672.
- [7] PARRILLI S,PODERICO M,ANGELINO C V,et al.A nonlocal SAR image denoising algorithm based on LLMMSE wavelet shrinkage[J].IEEE Transactions on Geoscience and Remote Sensing,2012,50(2):606-616.
- [8] 马晓双，吴鹏海，刘诗雨，等.结合相似块匹配及线性最小均方误差滤波器的全极化雷达影像去噪[J].遥感学报，2018,22(4):570-580.(MA Xiaoshuang,WU Penghai,LIU Shiyu,et al.Polarimetric radar image despeckling by using similar patches matching and minimum mean square error filter[J].Journal of Remote Sensing,2018,22(4):570-580.)
- [9] AUBERT G,AUJOL J F.A variational approach to removing multiplicative noise[J].SIAM Journal on Applied Mathematics,2008,68(4):925-946.
- [10] ZHANG K,ZUO W,CHEN Y,et al.Beyond a Gaussian denoiser:residual learning of deep CNN for image denoising[J].IEEE Transactions on Image Processing,2017,26(7):3142-3155.
- [11] ZHANG K,ZUO W,GU S,et al.Learning deep CNN denoiser prior for image restoration[C]//IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Honolulu,HI,USA:IEEE,2017:2808-2817.
- [12] CHIERCHIA G,COZZOLINO D,POGGI G,et al.SAR image despeckling through convolutional neural networks[C]//IEEE International Geoscience and Remote Sensing Symposium(IGARSS).Fort Worth,TX,USA:IEEE,2017:5438-5441.
- [13] ZHANG Q,YUAN Q,LI J,et al.Learning a dilated residual network for SAR image despeckling[J].Remote Sensing,2018,10(2):196.
- [14] 钱满，张向阳，李仁昌.改进卷积神经网络SAR图像去噪算法[J].计算机工程与应用，2020,56(14):176-182.(QIAN Man,ZHANG Xiangyang,LI Renchang.Improved convolutional neural network for SAR image despeckling algorithm[J].Computer Engineering and Applications,2020,56(14):176-182.)
- [15] LEE J S,HOPPEL K.Noise modeling and estimation of remotely-sensed images[C]//12th Canadian Symposium on Remote Sensing Geoscience and Remote Sensing Symposium.Vancouver,Canada,Canada:IEEE,1989,2:1005-1008.
- [16] YU F,KOLTUN V.Multi-scale context aggregation by dilated convolutions[EB/OL].[ 2020-10-22].https://arxiv.org/pdf/1511.07122.pdf.
- [17] IOFFE S,SZEGEDY C.Batch normalization:Accelerating deep network training by reducing internal covariate shift[EB/OL].[2020-10-22].https://www.ixueshu.com/document/f8e66fe8224877ea318947a18e7f9 386.html.
- [18] HARA K,SAITO D,SHOUNO H.Analysis of function of rectified linear unit used in deep learning[C]//2015 International Joint Conference on Neural Networks(IJCNN).Killarney,Ireland:IEEE,2015:1-8.
- [19] HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Las Vegas,NV,USA:IEEE,2016:770-778.
- [20] YANG Y,NEWSAM S.Bag-of-visual-words and spatial extensions for land-use classification[C]//Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems -GIS’10.San Jose,California.New York:ACM Press,2010:270-279.
- [21] KINGMA D P,BA J.Adam:a method for stochastic optimization[EB/OL].[2020-10-22].https://arxiv.org/abs/1412.6980v8.