Graph Convolutional Networks-Based Super-Resolution Land Cover Mapping

doi:10.1109/JSTARS.2021.3100400

	Graph Convolutional Networks-Based Super-Resolution Land Cover Mapping
	Zhang, Xining 2,3; Ge, Yong 2,3; Ling, Feng 4; Chen, Jin 5; Chen, Yuehong 1; Jia, Yuanxin 6
刊名	IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
	2021
卷号	14 页码:7667-7681
关键词	Deep neural networks (DNNs) graph convolutional networks (GCNs) land cover subpixel super-resolution mapping (SRM).
ISSN号	1939-1404
DOI	10.1109/JSTARS.2021.3100400
通讯作者	Ge, Yong(gey@lreis.ac.cn) ; Chen, Jin(chenjin@bnu.edu.cn)
英文摘要	Super-resolution mapping (SRM) is an effective technology to solve the problem of mixed pixels because it can be used to generate fine-resolution land cover maps from coarse-resolution remote sensing images. Current methods based on deep neural networks have been successfully applied to SRM, as they can learn complex spatial patterns from training data. However, they lack the ability to learn structural information between adjacent land cover classes, which is vital in the reconstruction of spatial distribution. In this article, an SRM method based on graph convolutional networks (GCNs), named SRMGCN, is proposed to improve SRM results by capturing structure information on the graph. In SRMGCN, a supervised inductive learning strategy with mini-graphs as input is considered, which is an extension of the GCN framework. Furthermore, two operations are designed in terms of adjacency matrix construction and an information propagation rule to help reconstruct detailed information of geographical objects. Experiments on three datasets with different spatial resolutions demonstrate the qualitative and quantitative superiority of SRMGCN over three other popular SRM methods.
资助项目	National Natural Science Foundation for Distinguished Young Scholars of China[41725006]
WOS关键词	NEURAL-NETWORK ; PIXEL ; CLASSIFICATION ; PATTERNS
WOS研究方向	Engineering ; Physical Geography ; Remote Sensing ; Imaging Science & Photographic Technology
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000684698600009
资助机构	National Natural Science Foundation for Distinguished Young Scholars of China
内容类型	期刊论文
源URL	[http://ir.igsnrr.ac.cn/handle/311030/164613]
专题	中国科学院地理科学与资源研究所
通讯作者	Ge, Yong; Chen, Jin
作者单位	1.Hohai Univ, Coll Hydrol & Water Resources, Nanjing 210098, Peoples R China 2.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing 100101, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 4.Chinese Acad Sci, Innovat Acad Precis Measurement Sci & Technol, Key Lab Monitoring & Estimate Environm & Disaster, Wuhan 430077, Peoples R China 5.Beijing Normal Univ, State Key Lab Earth Surface Proc & Resource Ecol, Beijing 100875, Peoples R China 6.Acad Forest Inventory & Planning, Natl Forestry & Grassland Adm, Beijing 100714, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Xining,Ge, Yong,Ling, Feng,et al. Graph Convolutional Networks-Based Super-Resolution Land Cover Mapping[J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING,2021,14:7667-7681.
APA	Zhang, Xining,Ge, Yong,Ling, Feng,Chen, Jin,Chen, Yuehong,&Jia, Yuanxin.(2021).Graph Convolutional Networks-Based Super-Resolution Land Cover Mapping.IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING,14,7667-7681.
MLA	Zhang, Xining,et al."Graph Convolutional Networks-Based Super-Resolution Land Cover Mapping".IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 14(2021):7667-7681.