杨晓晗,田馨

• 1:东南大学交通学院

摘要(Abstract)：

针对现有InSAR相位解缠方法在低相干和高相位梯度条件下稳定性不足，以及基于卷积神经网络的解缠方法受限于局部感受野、难以充分建模相位全局关联特性的问题，提出一种结合相位滤波的深度学习网络。该网络以Pyramid Vision Transformer为骨干，引入注意力增强和多尺度特征融合机制，并分两阶段训练分别用于相位滤波和解缠。在滤波阶段训练网络以预测噪声抑制后的相位；在解缠阶段，将原始缠绕相位与滤波相位在通道维度融合输入，直接预测解缠相位。实验结果表明，该方法在模拟数据和真实Sentinel-1数据中均表现出较好的解缠精度与稳定性，在复杂地形和严重失相干区域具有一定的应用潜力。

关键词(KeyWords)： InSAR;相位解缠;Transformer;相位滤波;深度学习

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 杨晓晗,田馨

DOI: 10.16251/j.cnki.1009-2307.2026.11.008

参考文献(References)：

• [1] 王义梅，罗小军，于冰，等.郑州市地面沉降InSAR监测[J].测绘科学，2019,44(9):100-106.(WANG Yimei,LUO Xiaojun,YU Bing,et al.Monitoring ground subsidence in Zhengzhou with InSAR[J].Science of Surveying and Mapping,2019,44(9):100-106.)
• [2] 江桥，彭军还，杨红磊，等.改进小基线集技术的GB-InSAR铁路边坡监测[J].测绘科学，2017,42(12):140-145.(JIANG Qiao,PENG Junhuan,YANG Honglei,et al.GB-InSAR railway slope monitoring of improved small baseline subset technology[J].Science of Surveying and Mapping,2017,42(12):140-145.)
• [3] 高海英，赵争，章彭.时序InSAR的贵州地质灾害监测[J].测绘科学，2020,45(7):91-99.(GAO Haiying,ZHAO Zheng,ZHANG Peng.Geological hazards in Guizhou by using time series SAR interferometry[J].Science of Surveying and Mapping,2020,45(7):91-99.)
• [4] CHEN C W,ZEBKER H A.Network approaches to two-dimensional phase unwrapping:Intractability and two new algorithms:erratum[J].Journal of the Optical Society of America A,2001,18(5):1192.
• [5] 杨硕航，高建，张嘉怡.一种使用自注意模型的一步式InSAR相位解缠方法[J].测绘科学，2023,48(11):145-152.(YANG Shuohang,GAO Jian,ZHANG Jiayi.A one-step InSAR phase unwrapping method using self-attention model[J].Science of Surveying and Mapping,2023,48(11):145-152.)
• [6] 梁峰，谢先明，徐有邈，等.一种改进的U-Net相位解缠方法[J].遥感信息，2021,36(5):134-141.(LIANG Feng,XIE Xianming,XU Youmiao,et al.An improved U-Net phase unwrapping method[J].Remote Sensing Information,2021,36(5):134-141.)
• [7] CHEN Yu,WANG Shuai,GAO Yandong,et al.DMP-PUNet:a novel network for two-dimensional InSAR phase unwrapping under severe noise and complex fringes conditions[J].International Journal of Applied Earth Observation and Geoinformation,2025,139:104519.
• [8] 侯国政，谢先明，贾金国，等.基于MAPNet的深度学习相位解缠方法[J].遥感信息，2024,39(5):86-101.(HOU Guozheng,XIE Xianming,JIA Jinguo,et al.Phase unwrapping method based on multiple attending path network[J].Remote Sensing Information,2024,39(5):86-101.)
• [9] WU Zhipeng,WANG Teng,WANG Yingjie,et al.Deep learning for the detection and phase unwrapping of mining-induced deformation in large-scale interferograms[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60:5216318.
• [10] CHEN Bingqian,YANG Yu,ZHANG Lipeng,et al.A novel knowledge-learning coupling method for InSAR phase unwrapping of large surface displacements in coal mining areas[J].IEEE Transactions on Geoscience and Remote Sensing,2024,62:4416320.
• [11] ZHOU Lifan,YU Hanwen,LAN Yang.Deep convolutional neural network-based robust phase gradient estimation for two-dimensional phase unwrapping using SAR interferograms[J].IEEE Transactions on Geoscience and Remote Sensing,2020,58(7):4653-4665.
• [12] WANG Hai,HU Jun,FU Haiqiang,et al.A novel quality-guided two-dimensional InSAR phase unwrapping method via GAUNet[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2021,14:7840-7856.
• [13] ZHOU Lifan,YU Hanwen,LAN Yang,et al.Deep learning-based branch-cut method for InSAR two-dimensional phase unwrapping[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60:5209615.
• [14] WU Zhipeng,WANG Teng,WANG Yingjie,et al.Deep-learning-based phase discontinuity prediction for 2-D phase unwrapping of SAR interferograms[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60:5216516.
• [15] ZHONG Wenjie,LI Jia,LI Xin,et al.A discontinuity-guided two-dimensional phase unwrapping method for SAR interferograms[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2024,17:9309-9323.
• [16] MURDACA G,RUCCI A,PRATI C.Deep learning for InSAR phase filtering:An optimized framework for phase unwrapping[J].Remote Sensing,2022,14(19):4956.
• [17] YU Hanwen,YANG Tianxiang,ZHOU Lifan,et al.PDNet:a lightweight deep convolutional neural network for InSAR phase denoising[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60:5239309.
• [18] GAO Yandong,YAO Jiaqi,ZHOU Wei,et al.CBIPDNet:A novel method for InSAR deformation interferometric phase filtering using deep learning network[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2024,17:15806-15815.
• [19] ZHOU Lifan,YU Hanwen.InSAR-DLPU:A benchmark dataset for deep learning-based synthetic aperture radar interferometry phase unwrapping [Software and Data Sets][J].IEEE Geoscience and Remote Sensing Magazine,2024,12(2):118-124.
• [20] WANG Wenhai,XIE Enze,LI Xiang,et al.Pyramid vision transformer:a versatile backbone for dense prediction without convolutions[C]//2021 IEEE/CVF International Conference on Computer Vision (ICCV).Montreal,QC,Canada:IEEE,2022:548-558.
• [21] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas,NV,USA:IEEE,2016:770-778.
• [22] ZHAO Hang,GALLO O,FROSIO I,et al.Loss functions for image restoration with neural networks[J].IEEE Transactions on Computational Imaging,2017,3(1):47-57.