Research on Urban Water Body Extraction based on Transfer Learning of Three High-resolution Image Datasets

doi:10.11873/j.issn.1004-0323.2022.3.0731

Remote Sensing Technology and Application

2022, Vol. 37

Issue (3): 731-738 DOI: 10.11873/j.issn.1004-0323.2022.3.0731

Research on Urban Water Body Extraction based on Transfer Learning of Three High-resolution Image Datasets

Jiarui Shi¹(

),Qian Shen²(

),Hongchun Peng¹,Liwei Li²,Yue Yao²,Mingxiu Wang²,Ru Wang¹

^1.School of Marine Technology and Geomatics，Jiangsu Ocean Unversity，Lianyungang 222005，China
^2.Aerospace Information Research Institute，Chinese Academy of Sciences，Beijing 100094，China

Download:

HTML

PDF (4805KB)
Export: BibTeX | EndNote (RIS)

Abstract

Water extraction is an essential step for rare earth monitoring of urban water environment. Extraction of small water bodies in the city has now become a hot depth study in the field of remote sensing images. However， deep learning requires a large number of sample datasets as input， and images with different spatial resolutions often need to construct different datasets. If the spatial resolution of the images is not much different， the sample transfer learning model can be used to ensure accuracy and save time. In this paper， the U-Net image segmentation model is selected to perform sample transfer learning for images with three different spatial resolutions—0.5 m， 0.8 m and 2 m respectively. It is found that after three migration learning of 2 meters to 0.8 meters， 2 meters to 0.5 meters， and 0.8 meters to 0.5 meters， the corresponding evaluation indexes F1-score， MioU and Kappa of the extracted water body are all above 0.80. Under the premise of little difference in resolution， this method of extracting urban water bodies from lower-resolution samples to higher-resolution images is basically feasible， and the accuracy of the results is better. It is suitable for water extraction in water-deficient cities.

Key words: High-resolution remote sensing image Sample transfer learning U-Net

Received: 27 November 2020 Published: 25 August 2022

ZTFLH:

P332

Corresponding Authors: Qian Shen E-mail: 2018224033@jou.deu.cn;shenqian@radi.ac.cn

	Service
	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Jiarui Shi
	Qian Shen
	Hongchun Peng
	Liwei Li
	Yue Yao
	Mingxiu Wang
	Ru Wang

Cite this article:

Jiarui Shi,Qian Shen,Hongchun Peng,Liwei Li,Yue Yao,Mingxiu Wang,Ru Wang. Research on Urban Water Body Extraction based on Transfer Learning of Three High-resolution Image Datasets. Remote Sensing Technology and Application, 2022, 37(3): 731-738.

URL:

http://www.rsta.ac.cn/EN/10.11873/j.issn.1004-0323.2022.3.0731 OR http://www.rsta.ac.cn/EN/Y2022/V37/I3/731

Fig.1 Model construction

Table 1 Remote sensing image parameters

Fig.2 High-resolution urban water body sample sets and raster label sample sets

Fig.3 Datasets division

Fig.4 U-net model structure

Fig.5 Improved Unet accuracy comparison

Fig.6 Five Model extract result indicators

Fig.7 Worldview-2 extract Small water body display

Fig.8 High-resolution image extraction results

Fig.9 Water extraction details display

Fig.10 Shenyang city water extraction results

Fig.11 Ningbo city water extraction results

1	Duan Hongtao， Luo Juhua， Cao Zhigang，et al.Progress in remote sensing of aquatic environments at the watershed scale［J］. Progress in Geography，2019，38（8）：1182-1195.
1	段洪涛，罗菊花，曹志刚，等.流域水环境遥感研究进展与思考［J］.地理科学进展，2019，38（8）：1182-1195.
2	Zhang Bing， Li Junsheng， Shen Qianet al.Key technologies and systems of surface water environment monitoring by remote sensing［J］. Environmental Monitoring in China，2019，35（4）：1-9.
2	张兵，李俊生，申茜，等.地表水环境遥感监测关键技术与系统［J］.中国环境监测，2019，35（4）：1-9.
3	Li L， Yan Z， Shen Q， et al. Water body extraction from very high spatial resolution remote sensing data based on fully convolutional networks［J］. Remote Sensing， 2019， 11（10）：1162-1181. DOI： . doi: 10.3390/rs11101162
4	Miao Z， Fu K， Sun H， et al. Automatic water-body segmentation from high-resolution satellite images via deep networks［J］. IEEE Geoence Science and Remote Sensing Letters， 2018：1-5. DOI： . doi: 10.1109/LGRS.2018.2794545
5	Isikdogan F， Bovik A C， Passalacqua P. Surface sater mapping by deep learning［J］. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing，2017，10（11）：4909-4918. DOI： . doi: 10.1109/JSTARS.2017.2735443
6	Ronneberger O， Fischer P， Brox T. U-Net： Convolutional networks for biomedical image segmentation［C］∥ International Conference on Medical Image Computing and Computer-assisted Intervention.Springer，Cham，2015：234-241.DOI： . doi: 10.1007/978-3-319-24574-4_28
7	Xu Huimin. Method research of high resolution remote sensing imagery classification based on U-Net model of deep learning［D］. Chengdu： Southwest Jiaotong University，2018.
7	许慧敏.基于深度学习U-Net模型的高分辨率遥感影像分类方法研究［D］. 成都：西南交通大学，2018.
8	Feng W， Sui H， Huang W， et al. Water body extraction from very high-resolution remote sensing imagery using deep U-Net and a superpixel-based conditional random field model［J］. IEEE Geoscience and Remote Sensing Letters，2018，16（4）：618-622. DOI： . doi: 10.1109/LGRS. 2018. 2879492
9	Ren Xinlei， Wang Yangpin， Yang Jingyu， et al. Building detection from remote sensing images based on improved U-Net［J］.Laser & Optoelectronics Progress，2019，56（22）：195-202.
9	任欣磊，王阳萍，杨景玉，等.基于改进U-Net的遥感影像建筑物提取［J］. 激光与光电子学进展，2019，56（22）：195-202.
10	Wang Zhuo， Yan Haowen， Lu Xiaomin，et al. High-resolution remote sensing image road extraction method for improving U-Ne［J］.Remote Sensing Technology and Application，2020，35（4）：741-748.
10	王卓，闫浩文，禄小敏，等. 一种改进 U-Net 的高分辨率遥感影像道路提取方法［J］. 遥感技术与应用，2020，35（4）：741-748.
11	Zhang Haoran， Zhao Jianghong， Zhang Xiaoguang. High-resolution image building extraction using U-Net neural network［J］. Remote Sensing Information，2020，35（3）：143-150.
11	张浩然，赵江洪，张晓光.利用U-Net网络的高分遥感影像建筑提取方法［J］.遥感信息，2020，35（3）：143-150.
12	Chen Y， Fan R， Yang X， et al. Extraction of urban water bodies from high-resolution remote sensing imagery using deep learning［J］. Water，2018，10（5）：585-605. DOI： . doi: 10.3390/ w10050585
13	Chen Y， Tang L， Kan Z， et al. A novel Water Body Extraction Neural Network （WBE-NN） for optical high-resolution multispectral imagery［J］.Journal of Hydrology，2020：125092-125103. DOI： . doi: 10.1016/j.jhydrol.2020.125092
14	Wang Ning， Cheng Jiahua， Zhang Hanye， et al. Application of U-Net Model to water extraction with high resolution remote sensing data［J］. Remote Sensing for Land & Resources，2020， 32（1）： 35-42.
14	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.
15	Pan S J， Yang Q. A survey on transfer learning［J］. IEEE Transactions on Knowledge and Data Engineering， 2010， 22（10）：1345-1359. DOI： . doi: 10.1109/TKDE.2009.191
16	Zhong C， Ting Z， Chao O. End-to-End airplane detection using transfer learning in remote sensing images［J］. Remote Sensing， 2018， 10（1）：139-164. DOI： . doi: 10.3390/rs10010139
17	Kemker R， Salvaggio C， Kanan C. Algorithms for semantic segmentation of multispectral remote sensing imagery using deep learning［J］. ISPRS Journal of Photogrammetry and Remote Sensing，2018，145：60-77. DOI： . doi: 10.1016/j.isprsjprs. 2018.04.014
18	Sun Xian， Liang Wei， Diao Wenhui， et al. Progress and challenges of remote sensing edge intelligence technology［J］. Journal of Image and Graphics，2020，25（9）：1719-1738.
18	孙显，梁伟，刁文辉，等. 遥感边缘智能技术研究进展及挑战［J］. 中国图象图形学报，2020，25（9）：1719-1738.
19	Chen Fu. Pixel Knife High-resolution Satellite Processing Software［EB/OL］. ，2020-06-07，2020-10-10.
19	陈甫.像素刻刀高分卫星处理软件［EB/OL］.，2020-06-07，2020-10-10.
20	Long T F， Jiao W L， He G J， et al. A fast and reliable matching method for automated georeferencing of remotely-sensed imagery［J］. Remote Sensing， 2016， 8（1）： 56–79. DOI： . doi: 10.3390/rs8010056
21	Guo Lifeng， Gao Xiaohong， Kang Jian， et al. Application of the pseudo-invariant feature in normalization process of the remote sensing images［J］. Remote Sensing Technology and Application， 2009，24（5）：588-595.
21	郭丽峰，高小红，亢健，等.伪不变特征法在遥感影像归一化处理中的应用［J］.遥感技术与应用，2009，24（5）：588-595.
22	Canty M J， Nielsen A A. Automatic radiometric normalization of multitemporal satellite imagery with the iteratively re-weighted MAD transformation［J］. Remote Sensing of Environment，2008.112（3）：1025-1036. DOI： . doi: 10.1016/j.rse. 2007. 07.013
23	Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition[EB/OL],,2015-5-10,2020-10-10.
24	He K， Zhang X， Ren S， et al. Deep residual learning for image recognition［C］∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas， NV， 2016： 770-778. DOI： . doi: 10.1109/CVPR.2016.90
25	Howard A G, Zhu M, Chen B, et al. Mobilenets: Efficient Convolutional Neural Networks for Mobile Vision Applications[EB/OL],,2017-4-17,2020-10-10.

No Suggested Reading articles found!

Viewed

Full text

Abstract

Cited

Shared

Discussed