基于改进U-Net网络的多源遥感影像洪涝灾害信息提取与变化分析

doi:10.13474/j.cnki.11-2246.2022.0005

测绘通报 ›› 2022, Vol. 0 ›› Issue (1): 26-32.doi: 10.13474/j.cnki.11-2246.2022.0005

• 第八届测绘科学前沿技术论坛获奖论文 • 上一篇下一篇

基于改进U-Net网络的多源遥感影像洪涝灾害信息提取与变化分析

车子杰, 高飞, 吴兆福, 李振轩

合肥工业大学土木与水利工程学院, 安徽合肥 230000

收稿日期:2021-06-11 发布日期:2022-02-22
通讯作者: 李振轩。E-mail:zxli2019@hfut.edu.cn
作者简介:车子杰(1997-),男,硕士生,研究方向为遥感影像信息提取与变化检测。E-mail:zijie_che@163.com
基金资助:
安徽省测绘科技专项资金（CHZX201801）；中央高校基本科研业务费专项资金（JZ2021HGTA0167）

Multi-source remote sensing image flood disaster information extraction and change analysis based on improved U-Net network

CHE Zijie, GAO Fei, WU Zhaofu, LI Zhenxuan

College of Civil Engineering, Hefei University of Technology, Hefei 230000, China

Received:2021-06-11 Published:2022-02-22

摘要/Abstract

摘要： 基于遥感影像的信息提取与变化分析是检测地球表面变化的重要方法，在城市规划、环境监测、灾害评估等方面应用广泛。合成孔径雷达（SAR）可以穿透云层，实现洪水期的对地观测，但侧视成像会随地形起伏产生叠掩、阴影等现象。DEM可以提供地面点的高程信息，反映地形的起伏变化，但无法提供地物信息。因此，本文提出了一种融合多源遥感影像、基于改进U-Net网络的变化检测方法；并通过试验证明了该方法可以实现高精度、高效的洪涝灾害信息提取。

关键词: U-Net, 深度学习, 变化检测, 数据增强, 高分三号

Abstract: Information extraction and change analysis based on remote sensing images is an important method for detecting changes on the earth's surface, and it has a wide range of applications in urban planning, environmental monitoring, disaster assessment. SAR can penetrate clouds to achieve ground observation during floods, but side-view imaging will cause overlaps and shadows along with terrain undulations. DEM can provide elevation information of ground points, reflecting the undulations of the terrain, but cannot provide surface information. In order to achieve high-precision and high-efficiency extraction of disaster information for the disaster area and degree of the flood area, this paper proposes a change detection method based on U-Net fusion of multi-source remote sensing images. Experimental results show that disaster flood information with high spatial resolution can be extracted efficiently by the method performed in this paper.

Key words: U-Net, deep learning, change detection, data enhancement, GF-3

中图分类号:

P237

车子杰, 高飞, 吴兆福, 李振轩. 基于改进U-Net网络的多源遥感影像洪涝灾害信息提取与变化分析[J]. 测绘通报, 2022, 0(1): 26-32.

CHE Zijie, GAO Fei, WU Zhaofu, LI Zhenxuan. Multi-source remote sensing image flood disaster information extraction and change analysis based on improved U-Net network[J]. Bulletin of Surveying and Mapping, 2022, 0(1): 26-32.

参考文献

[1] SHENG Y, GONG P, XIAO Q. Quantitative dynamic flood monitoring with NOAA AVHRR[J]. International Journal of Remote Sensing, 2001, 22(9):1709-1724.
[2] 马国斌,蒋卫国,李京,等.中国短时洪涝灾害危险性评估与验证[J].地理研究,2012,31(1):34-44.
[3] 李加林,曹罗丹,浦瑞良.洪涝灾害遥感监测评估研究综述[J].水利学报,2014,45(3):253-260.
[4] 张桂香,霍治国,吴立,等.1961-2010年长江中下游地区农业洪涝灾害时空变化[J].地理研究,2015,34(6):1097-1108.
[5] 李景刚, 黄诗峰, 李纪人. ENVISAT卫星先进合成孔径雷达数据水体提取研究——改进的最大类间方差阈值法[J]. 自然灾害学报, 2010, 19(3):139-145.
[6] HONG S, JANG H, KIM N, et al. Water area extraction using RADARSAT SAR imagery combined with Landsat imagery and terrain information[J].Sensors,2015, 15(3):6652-6667.
[7] 杜敬. 基于深度学习的湖泊湿地信息提取及时空演变特征研究[D].抚州:东华理工大学, 2017.
[8] LÄNGKVIST M, KISELEV A, ALIREZAIE M, et al. Classification and segmentation of satellite orthoimagery using convolutional neural networks[J]. Remote Sensing, 2016, 8(4):329.
[9] 王雪, 隋立春, 钟棉卿, 等. 全卷积神经网络用于遥感影像水体提取[J]. 测绘通报, 2018(6):41-45.
[10] WANG Z B, GAO X, ZHANG Y N, et al. MSLWENet:a novel deep learning network for lake water body extraction of google remote sensing images[J]. Remote Sensing, 2020, 12(24):4140.
[11] ZHANG H, GONG M G, ZHANG P Z, et al. Feature-level change detection using deep representation and feature change analysis for multispectral imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(11):1666-1670.
[12] SCOTT G J, ENGLAND M R, STARMS W A, et al. Training deep convolutional neural networks for land-cover classification of high-resolution imagery[J].IEEE Geoscience and Remote Sensing Letters, 2017,14(4):549-553.
[13] QI L, YONG D, XIN N, et al. Remote sensing image classification based on DBN model[J]. Journal of computer research and development, 2014, 51(9):1911.
[14] LI W J, FU H H, YU L, et al. Stacked Autoencoder-based deep learning for remote-sensing image classification:a case study of African land-cover mapping[J]. International Journal of Remote Sensing, 2016, 37(23):5632-5646.
[15] GONG M G, ZHAN T, ZHANG P Z, et al. Superpixel-based difference representation learning for change detection in multispectral remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(5):2658-2673.
[16] GENG J, WANG H Y, FAN J C, et al. Change detection of SAR images based on supervised contractive autoencoders and fuzzy clustering[C]//Proceedings of 2017 International Workshop on Remote Sensing with Intelligent Processing (RSIP).[S.l.]:IEEE,2017.
[17] 徐锐, 余小于, 张驰,等. 融合Unet网络和IR-MAD的建筑物变化检测方法[J]. 国土资源遥感, 2020, 32(4):90-96.
[18] RONNEBERGER O, FISCHER P, BROX T. U-net:Convolutional networks for biomedical image segmentation[C]//Proceedings of 2015 International Conference on Medical Image Computing and Computer-assisted Intervention.[S.l.]:Springer, 2015.
[19] 王宁, 程家骅, 张寒野,等. U-net模型在高分辨率遥感影像水体提取中的应用[J]. 国土资源遥感, 2020, 32(1):35-42.
[20] CHOLLET F. Xception:deep learning with depthwise separable convolutions[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition.[S.l.]:IEEE,2017.

基于改进U-Net网络的多源遥感影像洪涝灾害信息提取与变化分析

Multi-source remote sensing image flood disaster information extraction and change analysis based on improved U-Net network

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]	刘立, 董先敏, 刘娟, 文学虎. 人机融合智能的遥感解译生产新方法[J]. 测绘通报, 2022, 0(7): 118-123,137.
[2]	蔡柔丹. 一种基于用户异步轨迹的身份识别智能方法[J]. 测绘通报, 2022, 0(7): 158-162,167.
[3]	滕文鑫, 张建辰, 邵杰. 集成时空邻近与卷积网络车道级高精度定位算法[J]. 测绘通报, 2022, 0(6): 1-5,61.
[4]	窦世卿, 郑贺刚, 徐勇, 陈治宇, 苗林林, 宋莹莹. 基于U-Net3+的高分遥感影像建筑物提取[J]. 测绘通报, 2022, 0(6): 40-44.
[5]	张晋赫, 秦育罗, 张在岩, 宋伟东, 朱洪波. 复杂场景下农村道路裂缝分割方法[J]. 测绘通报, 2022, 0(5): 74-78,88.
[6]	阮明浩, 卢永华, 刘玉贤, 汤圣君, 王伟玺, 张宇琦, 李游. 激光与倾斜点云精确配准与无缝融合[J]. 测绘通报, 2022, 0(4): 6-10.
[7]	师芸, 石龙龙, 牛敏杰, 赵侃. 基于单阶段实例分割网络的黄土滑坡多任务自动识别[J]. 测绘通报, 2022, 0(4): 26-31.
[8]	王雪, 梁珂, 隋立春, 钟棉卿, 朱剑锋. 膨胀卷积与金字塔表达的深度学习模型用于农村建筑物提取[J]. 测绘通报, 2022, 0(4): 61-65.
[9]	潘建平, 李鑫, 孙博文, 胡勇, 李明明. 基于注意力密集连接金字塔网络的新增建设用地变化检测[J]. 测绘通报, 2022, 0(3): 41-46,59.
[10]	高鸣, 周鑫鑫, 刘琦, 杨光迪, 吴长彬. 深度学习网络支持下的农房侵占耕地自动化监测[J]. 测绘通报, 2022, 0(3): 47-53.
[11]	段晨阳, 冯建中, 全斌, 白林燕, 王盼盼. 利用深度学习进行GF-6影像枣园检测识别[J]. 测绘通报, 2022, 0(3): 54-59.
[12]	杨景玉, 贾多科, 王阳萍. 基于超像素分割与CRF的高分辨率遥感影像变化检测[J]. 测绘通报, 2022, 0(2): 25-30,49.
[13]	乐颖, 夏元平. 基于比值法的多源多时相数据城区变化检测[J]. 测绘通报, 2022, 0(2): 56-61.
[14]	孙琴琴, 蔡国印, 杨柳忠, 张宁, 杜明义. 城市尺度典型地表要素综合提取方法研究[J]. 测绘通报, 2022, 0(2): 62-66.
[15]	张自强, 刘涛, 杜萍, 锁旭宏, 杨国林. 图残差神经网络支持下的建筑物群组模式分类[J]. 测绘通报, 2022, 0(1): 1-7.