基于改进U-Net网络的高分遥感影像水体提取

doi:10.12082/dqxxkx.2020.190622

地球信息科学学报 ›› 2020, Vol. 22 ›› Issue (10): 2010-2022.doi: 10.12082/dqxxkx.2020.190622

• 专栏：城乡生态环境综合监测 • 上一篇下一篇

基于改进U-Net网络的高分遥感影像水体提取

何红术¹^,²(), 黄晓霞¹^,^*(), 李红旮¹, 倪凌佳¹^,², 王新歌³, 陈崇³, 柳泽³

1.中国科学院空天信息创新研究院,北京 100094
2.中国科学院大学,北京 100049
3.住房和城乡建设部城乡规划管理中心,北京 100835

收稿日期:2019-10-24 修回日期:2020-02-16 出版日期:2020-10-25 发布日期:2020-12-25
作者简介:何红术(1993- ), 男,重庆人,硕士生,研究方向为深度学习遥感信息提取、地理信息系统。E-mail:hhs10010@163.com
基金资助:
国家重点研发计划项目(2017YFB0503905);国家自然科学基金项目(41971363)

Water Body Extraction of High Resolution Remote Sensing Image based on Improved U-Net Network

HE Hongshu¹^,²(), HUANG Xiaoxia¹^,^*(), LI Hongga¹, NI Lingjia¹^,², WANG Xinge³, CHEN Chong³, LIU Ze³

1. Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China
2. University of Chinese Academy of Sciences, Beijing 100049, China
3. Urban and Rural Planning Management Center of the Ministry of Housing and Urban-Rural Development,Beijing 100835, China

Received:2019-10-24 Revised:2020-02-16 Online:2020-10-25 Published:2020-12-25
Contact: HUANG Xiaoxia
Supported by:
National Key Research and Development Program of China(2017YFB0503905);National Natural Science Foundation of China(41971363)

摘要/Abstract

摘要：

传统基于光谱信息的水体提取未能考虑水体形状、纹理、大小、相邻关系等问题,且存在同物异谱、异物同谱现象,导致水体提取精度较低。而传统基于分类提取水体方法设计特征过程较为繁琐,且不能挖掘深度信息特征。因此,本文提出改进的U-Net网络语义分割方法,借鉴经典U-Net网络的解编码结构对网络进行改进：① 将VGG网络用于收缩路径以提取特征;② 在扩张路径中对低维特征信息进行加强,将收缩特征金字塔上一层的特征图与下一层对应扩张路径上的特征图进行融合,以提高提取结果分割精度;③ 在分类后处理中引入条件随机场,以将分割结果精细化。在保持相同训练集、验证集和测试集的情况下,分别用SegNet、经典U-Net网络和改进的U-Net网络做对照试验。试验结果表明,改进的U-Net网络结构在IoU、精准率和Kappa系数指标上均高于SegNet和经典U-Net网络,与SegNet相比,3项指标分别提升了10.5%、12.3%和0.14,与经典U-Net网络结果相比,各个指标分别提升了5.8%、4.4%和0.05。改进的网络水体提取结果较为完整,对小目标水体能够准确提取。改进的U-Net网络能够有效地实现水体提取任务。

关键词: U-Net, 水体提取, 高分遥感影像, 条件随机场, 图像分割, VGG16, 青岛, 西宁

Abstract:

There are two main methods of traditional water body extraction: a method based on spectral information and a method based on classification. Traditional water body extraction methods based on spectral information fail to take into account features such as water body shape, internal texture, water body size, and adjacent relations of water body. Also, there is a common phenomenon of “same object with different spectra and same spectrum with different objects”, which could result in low accuracy of water body extraction. Thus, the traditional methods that design features based on classification to extract water body is complex and impossible to capture the deep information of water body features. This paper proposed an improved U-Net network semantic segmentation method, which uses the de-encoding structure of the classic U-Net network to improve the network: ① Use the VGG network to shrink the path and increase the depth of the network to extract deep features of the water; ② Strengthen the low-dimensional feature information in the expansion path, fuse the feature map on the next layer of the shrinking feature pyramid with the feature map on the corresponding expansion path in the next layer, and enhance the model's low-dimensional feature information to improve the classification accuracy of the model; and ③ The Conditional Random Feld (CRF) was introduced in the post-classification process to refine the segmentation results and improve the segmentation accuracy. In the study of Qingdao area, SegNet, classic U-Net network, and improved U-Net network were selected as controlled experiments while maintaining the same training set, validation set, and test set. The test results show that the improved U-Net network structure performed better than SegNet and classic U-Net networks in terms of IoU, accuracy rate and Kappa coefficient. Compared with SegNet, the three indicators increased by 10.5%, 12.3%, and 0.14, respectively. Compared with the results of the classic U-Net network, each indicator increased by 5.8%, 4.4% and 0.05, respectively. The results demonstrate the effectiveness of the improved method in this paper. In addition, the method proposed in this paper has more advantages than the other two networks in the extraction of small targets in the test area, the completeness of water body extraction, the distinction between shadows and water bodies, and the accuracy of boundary segmentation. In order to verify the space-time scalability of the model, this paper chose western Qingdao and Xining, Qinghai as the verification areas. The verification results show that the water body extraction was good for areas similar to the geographical environment of the experimental area, and the effect of water body extraction needs to be further improved in places that differ greatly from the geographical environment of the experimental area. In general, the improved U-Net network can effectively achieve the task of water extraction.

Key words: U-Net, water extraction, high-resolution remote sensing image, conditional random field, image segmentation, VGG16, Qingdao, Xining

何红术, 黄晓霞, 李红旮, 倪凌佳, 王新歌, 陈崇, 柳泽. 基于改进U-Net网络的高分遥感影像水体提取[J]. 地球信息科学学报, 2020, 22(10): 2010-2022.DOI:10.12082/dqxxkx.2020.190622

HE Hongshu, HUANG Xiaoxia, LI Hongga, NI Lingjia, WANG Xinge, CHEN Chong, LIU Ze. Water Body Extraction of High Resolution Remote Sensing Image based on Improved U-Net Network[J]. Journal of Geo-information Science, 2020, 22(10): 2010-2022.DOI:10.12082/dqxxkx.2020.190622

图/表 16

图1

图2

表1

图3

图4

图5

表2

图6

表3

表4

图7

表5

表6

表7

表8

表9

参考文献 28

[1]	Li K, Wu S, Dai E, et al. Flood loss analysis and quantitative risk assessment in China[J]. Natural Hazards, 2012,63(2):737-760.
[2]	Bond N R, Lake P, Arthington A H. The impacts of drought on freshwater ecosystems:an Australian perspective[J]. Hydrobiologia, 2008,600(1):3-16.
[3]	Lake P S. Ecological effects of perturbation by drought in flowing waters[J]. Freshwater Biology, 2003,48(7):1161-1172.
[4]	吴赛, 张秋文. 基于MODIS遥感数据的水体提取方法及模型研究[J]. 计算机与数字工程, 2005(7):1-4.
	[ Wu S, Zhang Q W. Method and model of water body extraction based on remote sensing data of MODIS[J]. Computer & Digital Engineering, 2005(7):1-4. ]
[5]	李丹, 吴保生, 陈博伟, 等. 基于卫星遥感的水体信息提取研究进展与展望[J]. 清华大学学报(自然科学版), 2020,60(2):147-161.
	[ Li D, Wu B S, Chen B W, et al. Review of waterbody information extraction based on satellite remote sensing[J]. Journal of Tsinghua University (Science and Technology), 2020,60(2):147-161. ]
[6]	Mcfeeters S K. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features[J]. International Journal of Remote Sensing, 1996,17(7):1425-1432.
[7]	Xu H. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery[J]. International Journal of Remote Sensing, 2006,27(14):3025-3033.
[8]	都金康, 黄永胜, 冯学智, 等. SPOT卫星影像的水体提取方法及分类研究[J]. 遥感学报, 2001,5(3):214-219.
	[ Du J K, Huang Y S, Feng X Z, et al. Study on water bodies extraction and classification from SPOT image[J]. Journal of Remote Sensing, 2001,5(3):214-219. ]
[9]	王培培. 基于ETM影像的水体信息自动提取与分类研究[J]. 首都师范大学学报(自然科学版), 2009,30(6):75-79.
	[ Wang P P. Automatic extraction and classification of water body from ETM image[J]. Journal of Capital Normal University (Natural Science Edition), 2009,30(6):75-79. ]
[10]	胡德勇, 李京, 陈云浩, 等. 单波段单极化SAR图像水体和居民地信息提取方法研究[J]. 中国图象图形学报, 2008,13(2):257-263.
	[ Hu D Y, Li J, Chen Y H, et al. Water and settlement area extraction from single band single poplarization SAR images based on SVM method[J]. Journal of Images and Graphics, 2008,13(2):257-263. ]
[11]	Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006,313(5786):504-507.
[12]	计梦予, 袭肖明, 于治楼. 基于深度学习的语义分割方法综述[J]. 信息技术与信息化, 2017(10):137-140.
	[ Ji M Y, Xi X M, Yu Z L. A review of semantic segmentation based on deep learning[J]. Information Technology and Informatization, 2017(10):137-140. ]
[13]	Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015: 3431-3440.
[14]	Badrinarayanan V, Handa A, Cipolla R. Segnet: A deep convolutional encoder-decoder architecture for robust semantic pixel-wise labelling[J]. ArXiv Preprint ArXiv:1505.07293, 2015: 117-124.
[15]	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, 2015: 234-241.
[16]	Zheng S, Jayasumana S, Romera-Paredes B, et al. Conditional random fields as recurrent neural networks[C]. Proceedings of the IEEE International Conference on Computer Vision, 2015: 1529-1537.
[17]	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.
[18]	许玥. 基于改进Unet的遥感影像语义分割在地表水体变迁中的应用[D]. 重庆:重庆师范大学, 2019.
	[ Xu Y. Application of semantic segmentation of remote sensing image based on improved Unet in surface water change[D]. Chongqing: Chongqing Normal University, 2019. ]
[19]	陈前, 郑利娟, 李小娟, 等. 基于深度学习的高分遥感影像水体提取模型研究[J]. 地理与地理信息科学, 2019,35(4):43-49.
	[ Chen Q, Zheng L J, Li X J, et al. Water body extraction from highresolution satellite remote sensing images based on deep learning[J]. Geography and Geo-information Science, 2019,35(4):43-49. ]
[20]	Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409. 1556, 2014.
[21]	Li Y, Guo L, Rao J, et al. Road segmentation based on hybrid convolutional network for high-resolution visible remote sensing image[J]. IEEE Geoscience and Remote Sensing Letters, 2018,16(4):613-617.
[22]	Hu J, Li L, Lin Y, et al. A Comparison and strategy of semantic segmentation on remote sensing images[J]. arXiv preprint arXiv:1905. 10231, 2019.
[23]	Krähenbühl P, Koltun V. Efficient inference in fully connected crfs with gaussian edge potentials [C]. Advances in neural information processing systems, 2011: 109-117.
[24]	Chen L-C, Papandreou G, Kokkinos I, et al. Semantic image segmentation with deep convolutional nets and fully connected crfs[J]. arXiv preprint arXiv:1412.7062, 2014: 97-105.
[25]	刘懿兰, 黄晓霞, 李红旮, 等. 基于卷积神经网络与条件随机场方法提取乡镇非正规固体废弃物[J]. 地球信息科学学报, 2019,21(2):259-268.
	[ Liu Y L, Huang X X, Li H G, et al. Extraction of irregular solid waste in rural based on Convolutional Neural Network and conditional random field method[J]. Journal of Geo-information Science, 2019,21(2):259-268. ]
[26]	黄英来, 刘亚檀, 任洪娥. 基于全卷积神经网络的林木图像分割[J]. 计算机工程与应用, 2019,55(4):219-224.
	[ Huang Y L, Liu Y T, Ren H E. Segmentation of forest image based on Fully Convolutional Neural Network[J]. Computer Engineering and Applications, 2019,55(4):219-224. ]
[27]	王晨巍, 王晓君. 高分遥感卫星影像的预处理技术[J]. 电子技术与软件工程, 2016(24):122-123.
	[ Wang C W, Wang X J. Pre-process technology of high precision satellite photograph[J]. Electronic Technology & Software Engineering. 2016(24):122-123. ]
[28]	张新明, 祝晓斌, 蔡强, 等. 图像语义分割深度学习模型综述[J]. 高技术通讯, 2017,27(Z1):808-815.
	[ Zhang X M, Zhu X B, Cai Q, et al. Survey of the deep learning models for image semantic segmentation[J]. Chinese High Technology Letters, 2017,27(Z1):808-815. ]

	感受野	步长	填充	输出大小
InputRGBimage：3@256×256
Conv+ReLU	3×3	1	1	64@256×256
Conv+ReLU	3×3	1	1	64@256×256
Max-pooling				64@128×128
Conv+ReLU	3×3	1	1	128@128×128
Conv+ReLU	3×3	1	1	128@128×128
Max-pooling				128@64×64
Conv+ReLU	3×3	1	1	256@64×64
Conv+ReLU	3×3	1	1	256@64×64
Conv+ReLU	3×3	1	1	256@64×64
Max-pooling				256@32×32
Conv+ReLU	3×3	1	1	512@32×32
Conv+ReLU	3×3	1	1	512@32×32
Conv+ReLU	3×3	1	1	512@32×32
Max-pooling				512@16×16
Conv+ReLU	3×3	1	1	512@16×16
Conv+ReLU	3×3	1	1	512@16×16
Conv+ReLU	3×3	1	1	512@16×16
Max-pooling				512@8×8

影像编号	中心经度/°E	中心纬度/°N	成像时间	影像大小/像素×像素
L1A0003593712	120.5	36.7	2018-11-12	27 620×35 273
L1A0003593719	120.4	36.3	2018-11-12	27 620×35 113
L1A0003593868	120.6	36.3	2018-11-12	27 620×35 191

	实际正类	实际负类
预测正类	TP	FP
预测负类	FN	TN

方法	IoU/%	精准率/%	Kappa系数
SegNet	77.6	82.5	0.79
经典U-net	82.3	90.4	0.88
改进后的U-Net网络	88.1	94.8	0.93

影像编号	中心经度/°E	中心纬度/°N	成像时间	影像大小/像素×像素
L1A0003553729	120.1	36.3	2018-10-28	276 20×292 00
L1A0003351642	101.5	36.8	2018-07-26	276 20×292 00

基于改进U-Net网络的高分遥感影像水体提取

Water Body Extraction of High Resolution Remote Sensing Image based on Improved U-Net Network

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 28

相关文章 15

Metrics

本文评价

推荐阅读 0

		区域1	区域2	区域3	区域4	区域5
青岛	原始影像
	水体信息
西宁	原始影像
	水体信息

[1]	张旭, 柳林, 周翰林, 岳瀚, 孙秋远. 基于贝叶斯逻辑回归模型研究百度街景图像微观建成环境因素对街面犯罪的影响[J]. 地球信息科学学报, 2022, 24(8): 1488-1501.
[2]	刘佳, 伍宇明, 高星, 司文涛. 基于GEE和U-net模型的同震滑坡识别方法[J]. 地球信息科学学报, 2022, 24(7): 1275-1285.
[3]	姚锦一, 王卷乐, 严欣荣, 魏海硕, Altansukh Ochir, Davaadorj Davaasuren. 基于深度神经网络的蒙古国色楞格河流域水体信息提取[J]. 地球信息科学学报, 2022, 24(5): 1009-1017.
[4]	邱凤婷, 过志峰, 张宗科, 魏显虎, 景睦馨. 基于改进SVM分类法的SAR图像水体面积提取研究[J]. 地球信息科学学报, 2022, 24(5): 940-948.
[5]	陈振, 陈芸芝, 吴婷, 李佳优. 面向高分遥感影像道路提取的轻量级双注意力和特征补偿残差网络模型[J]. 地球信息科学学报, 2022, 24(5): 949-961.
[6]	林禹, 赵泉华, 李玉. 一种基于深度传递迁移学习的遥感影像分类方法[J]. 地球信息科学学报, 2022, 24(3): 495-507.
[7]	习文强, 杜世宏, 杜守基. 多时相耕地覆盖提取和变化分析：一种结合遥感和空间统计的时空上下文方法[J]. 地球信息科学学报, 2022, 24(2): 310-325.
[8]	张春森, 贾欣, 吴蓉蓉, 崔卫红, 史书, 郭丙轩. 面向对象高分遥感影像典型自然地物半自动提取[J]. 地球信息科学学报, 2021, 23(6): 1050-1062.
[9]	李志红, 李旺平, 王玉, 陈璐, 郁林, 周兆叶, 郝君明, 吴晓东, 李传华. 基于离散粒子群算法的2种新型水体提取方法的对比与验证[J]. 地球信息科学学报, 2021, 23(6): 1106-1117.
[10]	乔丹玉, 郑进辉, 鲁晗, 邓磊. 面向不同环境背景的Landsat影像水体提取方法适用性研究[J]. 地球信息科学学报, 2021, 23(4): 710-722.
[11]	朱盼盼, 李帅朋, 张立强, 李洋. 基于多任务学习的高分辨率遥感影像建筑提取[J]. 地球信息科学学报, 2021, 23(3): 514-523.
[12]	冯叶涵, 陈亮, 贺晓冬. 基于百度街景的SVF计算及其在城市热岛研究中的应用[J]. 地球信息科学学报, 2021, 23(11): 1998-2012.
[13]	赵青松, 兰安军, 范泽孟, 杨青. 贵州省不同地貌形态类型土壤侵蚀强度变化的定量分析[J]. 地球信息科学学报, 2020, 22(7): 1555-1566.
[14]	郭子慧, 刘伟. 深度学习和遥感影像支持的矢量图斑地类解译真实性检查方法[J]. 地球信息科学学报, 2020, 22(10): 2051-2061.
[15]	刘晓慧,刘永伟,蔡菲,樊文平. 面向时空可达性的城市应急医疗机构推荐方法[J]. 地球信息科学学报, 2019, 21(9): 1411-1419.