融合多尺度雷达后向散射特征深度学习的多极化SAR影像山区水体高精度提取

doi:10.13474/j.cnki.11-2246.2026.0201

测绘通报 ›› 2026, Vol. 0 ›› Issue (2): 1-6.doi: 10.13474/j.cnki.11-2246.2026.0201

• 水陆环境监测与资源评估 • 下一篇

融合多尺度雷达后向散射特征深度学习的多极化SAR影像山区水体高精度提取

原志芳¹, 李俊晓², 康倩²

1. 深圳大学土木与交通工程学院, 广东深圳 518000;
2. 山西玖立信息科技股份有限公司, 山西太原 030000

收稿日期:2025-08-07 发布日期:2026-03-12
通讯作者: 李俊晓。E-mail:1114630830@qq.com
作者简介:原志芳(1989—),女,硕士,高级工程师,主要研究方向为自然资源信息化。E-mail:sxjlxxkj@163.com
基金资助:
国家自然科学基金面上项目(42371431);国家重点研发计划(2022YFB3903705)

High-precision extraction of water bodies in mountainous areas from multi-polarization SAR images via deep learning that integrates multi-scale radar backscatter characteristics

YUAN Zhifang¹, LI Junxiao², KANG Qian²

1. College of Civil and Transportation Engineering, Shenzhen University, Shenzhen 518000, China;
2. Shanxi Jiuli Information Technology Co., Ltd., Taiyuan 030000, China

Received:2025-08-07 Published:2026-03-12

摘要/Abstract

摘要： 针对山区阴影、植被茂密和SAR相干斑噪声易导致水体误提和水体边界细节保持困难等问题,本文综合利用多极化SAR数据,提出了一种融合多尺度雷达后向散射特征深度学习的SAR影像水体提取方法。该方法采用多分辨率残差卷积和多尺度特征恒等映射来更好地表征山区复杂环境下多极化SAR影像丰富的多尺度地物特征,以提高水体提取的精度和完整性。试验采用覆盖山西省某水库的Sentinel-1双极化数据,并选择经典SVM、MRF算法和U-Net网络模型与本文算法进行水体提取性能的定性定量对比分析。F1、Kappa系数、IoU和OA分别达到92.14%、91.39%、85.43%和98.62%,表明本文方法水体提取综合性能最优,而且能够较好地保持水体提取的边界细节。

关键词: 水体提取, 多极化SAR, 深度学习, 多尺度, 山体阴影

Abstract: To address the challenges of frequent water misclassification and difficulty in preserving water boundary details caused by mountain shadows,dense vegetation,and SAR coherent speckle noise in complex terrains,this study proposes a novel deep learning-based water extraction method for SAR imagery that integrates multi-scale radar backscattering features through multi-polarization SAR data.The developed approach employs multi-resolution residual convolution and multi-scale feature identity mapping to effectively characterize the rich multi-scale terrain features in complex mountainous environments using multi-polarization SAR imagery,thereby enhancing both accuracy and completeness in water body extraction.Experimental validation was conducted using Sentinel-1 dual-polarization data covering a reservoir in Shanxi Province,with comparative analysis against classical SVM,MRF algorithm,and U-Net model through qualitative and quantitative assessments.The experimental results showed that 92.14% F1,91.39% Kappa coefficient,85.43% IoU,and 98.62% OA.This indicates that the method proposed in this paper has the best comprehensive performance in water extraction and can effectively maintain the boundary details of water extraction.

Key words: waterbody extraction, multi-polarization SAR, deep learning, multi-scale, mountain shadows

中图分类号:

P237

原志芳, 李俊晓, 康倩. 融合多尺度雷达后向散射特征深度学习的多极化SAR影像山区水体高精度提取[J]. 测绘通报, 2026, 0(2): 1-6.

YUAN Zhifang, LI Junxiao, KANG Qian. High-precision extraction of water bodies in mountainous areas from multi-polarization SAR images via deep learning that integrates multi-scale radar backscatter characteristics[J]. Bulletin of Surveying and Mapping, 2026, 0(2): 1-6.

参考文献

[1] 魏钜杰.复杂地形区域合成孔径雷达正射影像制作方法研究[D].阜新:辽宁工程技术大学,2009.
[2] 高寒新,陈波,孙洪泉,等.SAR卫星影像洪水检测研究进展及展望[J].地球信息科学学报,2023,25(10):1933-1953.
[3] 朱璨阳,耿君,李金超,等.基于Sentinel-1卫星SAR数据的2020年巢湖洪涝监测及灾情分析[J].合肥工业大学学报(自然科学版),2023,46(9):1217-1223.
[4] CHINI M,HOSTACHE R,GIUSTARINI L,et al.A hierarchical split-based approach for parametric thresholding of SAR images:flood inundation as a test case[J].IEEE Transactions on Geoscience and Remote Sensing,2017,55(12):6975-6988.
[5] 安成锦,牛照东,李志军,等.典型Otsu算法阈值比较及其SAR图像水域分割性能分析[J].电子与信息学报,2010,32(9):2215-2219.
[6] LI Junhua,WANG Shusen.An automatic method for mapping inland surface waterbodies with Radarsat-2 imagery[J].International Journal of Remote Sensing,2015,36(5):1367-1384.
[7] LOBRY S,DENIS L,TUPIN F,et al.Double MRF for water classification in SAR images by joint detection and reflectivity estimation[C]//Proceedings of 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).Fort Worth:IEEE,2017:2283-2286.
[8] XIE Cong,ZHANG Xin,ZHANG Xianyi,et al.Delineation of surface water bodies from SAR imagery based on improved MRF and CNN model[C]//Proceedings of the 8th International Conference on Image,Vision and Computing (ICIVC).Dalian:IEEE,2023:478-482.
[9] ZHENG Xinhong,CHEN Jiaqi,ZHANG Shilin,et al.Water extraction of SAR image based on region merging algorithm[C]//Proceedings of 2017 International Applied Computational Electromagnetics Society Symposium (ACES).Suzhou:IEEE,2017:1-2.
[10] LIU Zhongling,LI Fei,LI Ning,et al.A novel region-merging approach for coastline extraction from sentinel-1A IW mode SAR imagery[J].IEEE Geoscience and Remote Sensing Letters,2016,13(3):324-328.
[11] 李宁,牛世林.基于局部超分辨重建的高精度SAR图像水域分割方法[J].雷达学报,2020,9(1):174-184.
[12] 郑长利,黄国满,赵争,等.G0和Gamma联合建模的水平集水体分割[J].遥感信息,2018,33(1):13-20.
[13] WU Lin,WANG Le,MIN Lin,et al.Discrimination of algal-bloom using spaceborne SAR observations of great lakes in China[J].Remote Sensing,2018,10(5):767.
[14] QIN Xingli,YANG Jie,LI Pingxiang,et al.Research on water body extraction from Gaofen-3 imagery based on polarimetric decomposition and machine learning[C]//Proceedings of 2019 IEEE International Geoscience and Remote Sensing Symposium.Yokohama:IEEE,2019:6903-6906.
[15] 郭丽清,况润元.基于Sentinel-1 SAR数据的鄱阳湖水体提取方法对比研究[J].水文,2024,44(5):68-74.
[16] BOULZE H,KOROSOV A,BRAJARD J.Classification of sea ice types in Sentinel-1 SAR data using convolutional neural networks[J].Remote Sensing,2020,12(13):2165.
[17] 徐康,朱茂,贺秋华,等.一种基于TransUNet的SAR影像水体提取及洲滩面积变化监测应用[J].测绘科学,2024,49(2):55-64.
[18] WEI Jujie,ZHANG Yonghong,WU Hong'an,et al.An efficient change detection for large SAR images based on modified U-Net framework[J].Canadian Journal of Remote Sensing,2020,46(3):272-294.
[19] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al.Delving deep into rectifiers:surpassing human-level performance on ImageNet classification[C]//Proceedings of 2015 IEEE International Conference on Computer Vision (ICCV).Santiago:IEEE,2015:1026-1034.

融合多尺度雷达后向散射特征深度学习的多极化SAR影像山区水体高精度提取

High-precision extraction of water bodies in mountainous areas from multi-polarization SAR images via deep learning that integrates multi-scale radar backscatter characteristics

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	纪硕磊, 黄恒英, 李宇程, 陈海林. 基于多尺度特征融合的无人机紫外成像220 kV高压设备局部放电检测[J]. 测绘通报, 2026, 0(2): 161-167.
[2]	林羊, 于冰, 田馨, 张冠军, 刘成, 甘俊, 李广宇. 京雄城际铁路沿线InSAR地表沉降监测与影响因素分析[J]. 测绘通报, 2026, 0(2): 46-53.
[3]	张欢, 黄秋影, 刘胜, 王奇, 李凯. 基于全局和局部特征融合的作物细粒度分类算法[J]. 测绘通报, 2026, 0(2): 144-148.
[4]	叶元元, 陈春晖, 厉香蕴, 方婷婷, 李金超. 基于SNUNet的国产卫星影像耕地变化检测[J]. 测绘通报, 2026, 0(1): 130-134.
[5]	刘丹莹, 夏既胜. 基于改进PSPNet网络的林窗信息提取方法[J]. 测绘通报, 2026, 0(1): 151-155,171.
[6]	赵慧, 刘茜, 张敏, 李佳玉. 基于MGWR模型的新疆景观格局与防风固沙服务关系分析[J]. 测绘通报, 2025, 0(9): 1-7.
[7]	杨颜光, 钱建国, 于斌, 郭洁, 焦扬. 轻量的增强型激光雷达-惯性-视觉里程计系统[J]. 测绘通报, 2025, 0(9): 78-83,104.
[8]	江峰, 陈超. 基于深度学习的耕地信息提取与地表变化监测分析[J]. 测绘通报, 2025, 0(9): 131-134.
[9]	徐雯婷, 闫冬梅, 王虎, 吴雅睿, 王美景, 申茜. 水体指数结合DEM自适应搜索算法的线状水体提取方法[J]. 测绘通报, 2025, 0(9): 39-44,77.
[10]	唐芝青, 张涛, 王培玉, 向导, 刘海飞, 刘仁峰, 贺江江. ADM-YOLOv11:动态自适应铁塔视频多尺度目标检测算法[J]. 测绘通报, 2025, 0(9): 70-77.
[11]	邓小龙, 黄志海, 郭波. 改进分割大模型的桥梁缆索损伤语义分割方法[J]. 测绘通报, 2025, 0(9): 112-117.
[12]	汪志刚, 冯凯, 刘懿晨, 刘辉, 张文俊. 融合LightGlue的双目视觉SLAM算法[J]. 测绘通报, 2025, 0(9): 168-172.
[13]	胡耀锋, 程相兵, 陈嘉琦. 城市立体感知网建设关键技术与监测应用——以广州市历史文化街区为例[J]. 测绘通报, 2025, 0(8): 174-178.
[14]	李超, 刘清屹, 张佳伟, 石勇, 杨敏. 基于图像增强和改进RT-DETR的水下垃圾检测算法[J]. 测绘通报, 2025, 0(8): 128-136.
[15]	陈登峰, 赵航辉, 刘世鹏, 孟屯良, 王泽鹏. 集成边缘门控与多尺度空间注意的排水管道缺陷分割模型[J]. 测绘通报, 2025, 0(6): 97-102.