基于语义分割网络的GF-7 号卫星数据水体信息提取

大坝与安全 ›› 2022 ›› Issue (4): 50-.

基于语义分割网络的GF-7 号卫星数据水体信息提取

李红林¹，隋百凯²，叶燕萍¹，曹云刚²

1. 浙江华东测绘与工程安全技术有限公司，浙江杭州，310003；2. 西南交通大学地球科学与环境工程学院，四川成都，611756

收稿日期:2021-12-08 出版日期:2022-08-08 发布日期:2022-10-21
作者简介:李红林（1986— ），男，山东泰安人，高级工程师，研究方向为摄影测量与遥感。
基金资助:
中国电力建设集团股份有限公司科技项目（DJZDXM-2019-36）

Water information extraction from GF-7 satellite data based on semantic segmentation network

LI Honglin, SUI Baikai, YE Yanping and CAO Yungang

Zhejiang Huadong Mapping and Engineering Safety Technology Co., Ltd.

Received:2021-12-08 Online:2022-08-08 Published:2022-10-21

摘要/Abstract

摘要： 水体信息提取是水资源管理与保护的重要前提。遥感技术是水体提取的一个重要手段，现有的水体提取方法大多应用于中低分辨率遥感影像，而对高分辨率遥感影像，尤其是GF-7卫星影像的水体提取研究较少。笔者针对GF-7数据空间分辨率高和像素级数庞大等特点，提出了一个基于语义分割网络的水体信息自动提取方法。在该研究中，基于GF-7的立体像对数据进行数字地表模型（DSM）的提取，并通过波段组合的形式进行学习训练，有效缓解了复杂地形或背景下的水体错分和漏分现象。网络使用空洞卷积和空间金字塔池化结构有效提高了模型大尺度空间特征提取和多空间尺度下的语义信息学习能力。以湖北省泉河流域为研究区，进行了水体提取实验及不同方法的对比实验。由于影像非水体信息较多，引入平均交并比MIOU 作为评价指标。实验结果表明，该方法可有效实现GF-7影像水体信息的高精度提取，与传统水体指数法、支持向量机及U-Net方法相比，提取精度显著提高，水体提取精度可达到95.3%，MIOU 高达0.932。

关键词: 水体信息提取, 卫星遥感, GF-7号, 语义分割网络

Abstract: Water information extraction is an important prerequisite for water resources management and protection. Remote sensing technology is an important means of water extraction. Most of the exist. ing water extraction methods are applied to low and medium resolution remote sensing images, while there is little research on water extraction from high-resolution remote sensing images, especially GF-7 satellite images. Combined with the characteristics of high spatial resolution and huge pixel number of GF-7 data, an automatic water information extraction method based on semantic segmentation network is proposed in this paper. In this study, the digital surface model DSM is extracted from the data based on the stereo image of GF-7, and learning and training in the form of band combination are carried out, which effectively alleviates the water misclassification and missing in complex terrain or background. The network uses hole convolution and spatial pyramid pool structure, which effectively improves the ability of large-scale spatial feature extraction and semantic information learning at multiple spatial scales. Taking Quanhe river basin in Hubei province as the study area, water extraction experiments and comparative experiments of different methods are carried out. Due to the large amount of non-water information in the image, the average intersection ratio MIOU is introduced as the evaluation index. The experiment results show that this method can effectively realize the high-precision extraction of water information from GF-7 image, and the extraction accuracy is significantly improved compared with the traditional water index method, support vector machine and U-Net method. The water extraction accuracy can reach 95.3% and MIOU can reach 0.932.

Key words: water information extraction, satellite remote sensing, GF-7, semantic segmentation network

中图分类号:

TV211

李红林, 隋百凯, 叶燕萍, 曹云刚. 基于语义分割网络的GF-7 号卫星数据水体信息提取[J]. 大坝与安全, 2022, 0(4): 50-.

LI Honglin, SUI Baikai, YE Yanping and CAO Yungang. Water information extraction from GF-7 satellite data based on semantic segmentation network[J]. Dam & Safety, 2022, 0(4): 50-.

参考文献

[1] 吴文渊,沈晓华,邹乐君,等.基于Landsat ETM +影像的水体信息综合提取方法[J].科技通报,2008,24(2):252-259.
[2] 丁凤.基于新型水体指数（NWI）进行水体信息提取的实验研究[J].测绘科学,2009,34(4):155-157.
[3] XU H Q.A Study on Information Extraction of Water Body with the Modified Normalized Difference Water Index (MNDWI)[J].Journal of Remote Sensing,2005,9(5):589-595.
[4] 程晨,韦玉春,牛志春.基于ETM+图像和决策树的水体信息提取——以鄱阳湖周边区域为例[J].遥感信息,2012,27(6):49-56.
[5] Xia X R,Wei Y C,Xu N,et al.Decision tree model of extracting blue-green algal blooms information based on Landsat TM/ETM+imagery in Lake Taihu[J].Journal of Lake sciences,2014,26(6):907-915.
[6] 岁秀珍,阮仁宗,马荣华,等.面向对象的多尺度水体信息提取[J].地理空间信息,2012(6):72-75.
[7] 白晓燕,陈晓宏,王兆礼.基于面向对象分类的土地利用信息提取及其时空变化研究[J].遥感技术与应用,2015,30(4):798-809.
[8] 杨莹,阮仁宗.基于TM影像的平原湖泊水体信息提取的研究[J].遥感信息,2010(3):60-64.
[9] 刘家福,李壮,张昱鑫,等.基于几何特征与NDWI的城市水体信息提取研究[J].吉林师范大学学报（自然科学版）2019,40(1):121-127.
[10] T. D. Acharya,D H Lee,I T Yang,et al.Identification of Water Bodies in a Landsat 8 OLI Image Using a J48 Decision Tree[J].Sensors,2016,16(7):1075.
[11] 童李霞,燕琴,骆成凤,等.基于NDWI分割与面向对象的水体信息提取[J].地理空间信息,2017,15(5):57-59.
[12] 段秋亚,孟令奎,樊志伟,等.GF-1卫星影像水体信息提取方法的适用性研究[J].国土资源遥感,2015,27(4):79-84.
[13] 赵慧童,王萍,张娜,等.基于GF-1遥感数据的城市水体信息提取方法研究[J]. 测绘与空间地理信息,2017,40(10):70-73,76.
[14] 邹橙,杨学志,董张玉,等.基于GF-2遥感影像的一种快速水体信息提取方法[J].图学学报,2019,40(1):99-104.
[15] 谷鑫志,曾庆伟,谌华,等.高分三号影像水体信息提取[J].遥感学报,2019(3):555-565.
[16] E. Maggiori,Y. Tarabalka,G. Charpiat,et al.Convolutional Neural Networks for Large-Scale Remote Sensing Image Classification[J].IEEE Transactions on Geoscience & Remote Sensing,2016,55(2):645-657.
[17] Chen S W,Tao C S.PolSAR Image Classification Using Polarimetric- Feature- Driven Deep Convolutional Neural Network[J].IEEE Geoscience and Remote Sensing Letters,2018,15(4):627-631.
[18] Sui B K,Jiang T,Zhang Z,et al.A Modeling Method for Automatic Extraction of Offshore Aquaculture Zones Based
on Semantic Segmentation[J].International Journal of GeoInformation,2020,9(3):145.
[19] 何海清,杜敬,陈婷,等.结合水体指数与卷积神经网络的遥感水体提取[J].遥感信息,2017,32(5):82-86.
[20] 梁泽毓,吴艳兰,杨辉,等.基于密集连接全卷积神经网络的遥感影像水体全自动提取方法[J].遥感信息,2020,35(4):68-77.
[21] Qin P,Cai Y and Wang X L.Small Waterbody Extraction With Improved U-Net Using Zhuhai-1 Hyperspectral Remote Sensing Images[J].IEEE Geoscience and Remote Sensing Letters,2022(19):1-5.
[22] A. Krizhevsky,I. Sutskever,G. E. Hinton.ImageNet Classification with Deep Convolutional Neural Networks[C].Neural Information Processing Systems,2012.
[23] Miao Z H, Liu L K, Ren L,et al.Semantic Segmentation of Remote Sensing Images Based on Multi-Model Fusion[J].Journal of Physics: Conference Series,2020,1575(1):012119.
[24] C. Geiß, Zhu Y, Qiu C P,et al.Deep Relearning in the Geospatial Domain for Semantic Remote Sensing Image Segmentation[J].IEEE Geoscience and Remote Sensing Letters,2020,19:1-5.
[25] J. Long,E. Shelhamer,T. Darrell.Fully Convolutional Networks for Semantic Segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39(4):640-651.
[26] O. Ronneberger,P. Fischer,T. Brox.U-Net: Convolutional Networks for Biomedical Image Segmentation[C].International Conference on Medical Image Computing and Computer-Assisted Intervention,2015.
[27] V. Badrinarayanan,A. Kendall,R. Cipolla.SegNet: A Deep Convolutional Encoder- Decoder Architecture for Image Segmentation[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(12):2481-2495.
[28] Chen L C, Zhu Y,G. Papandreou,et al.Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation[C].Proceedings of the European conference on computer vision (ECCV),2018:801-818.
[29] Wu Z F, Shen C H,A. V. Hengel.Wider or Deeper: Revisiting the ResNet Model for Visual Recognition[J].Pattern Recognition,2019,90:119-133.
[30] Chen L C,G. Papandreou,I. Kokkinos,et al.DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(4):834-848.
[31] S. Ioffe,C. Szegedy.Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift[C].Proceedings of the International Conference on Machine Learning,2015.
[32] D. P. Kingma,J. Ba.Adam: A Method for Stochastic Optimization[J].Computer Science,2014:6628106.
[33] Y. Gonzalez,Shen C Y,H U Jung,et al.Semi-automatic sigmoid colon segmentation in CT for radiation therapy treatment planning via an iterative 2.5-D deep learning approach[J].Medical Image Analysis,2021,68:101896.
[34] G. E. Hinton,N. Srivastava,A. Krizhevsky,et al.Improving neural networks by preventing co- adaptation of feature detectors[J].Computer Science,2012,3(4):212-223.