High-Resolution Land Cover Mapping Through Learning With Noise Correction

doi:10.1109/TGRS.2021.3068280

	High-Resolution Land Cover Mapping Through Learning With Noise Correction
	Dong, Runmin 1,2; Fang, Weizhen 3; Fu, Haohuan 1; Gan, Lin 4; Wang, Jie 5; Gong, Peng 1
	2022
发表期刊	IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
ISSN	0196-2892
卷号	60 页码:13
摘要	High-resolution land cover mapping over large areas is a challenging task due to the lack of high-quality labels. A potential solution is to leverage the existing knowledge contained in the freely available lower-resolution land cover products. However, the relatively low resolution and low accuracy of the products lead to numerous inaccurate labels, which harms the performance of the neural network. This article addresses the challenge by jointly optimizing the network parameters and correcting the noisy labels with a novel online noise correction approach and a synergistic noise correction loss. By incorporating the information entropy as a measurement to determine the probable correct labels, the proposed noise correction approach learns to make effective correction of the noisy labels during training and eventually boosts the performance with a training set containing less noisy labels. Experimental results show that the proposed method can effectively correct the noisy labels and reduce their negative impact on network training. By employing the proposed method, we produce a refined high-resolution (3-m) land cover map from a lower-resolution (10-m) product in China and improve the accuracy from 74.96x0025; (10-m) to 81.32x0025; (3-m). Such an approach that can effectively learn from noisy data sets leads to many potential opportunities for using and magnifying existing knowledge and results.
关键词	Noise measurement Training Task analysis Image resolution Satellites Noise robustness Image segmentation Deep learning high-resolution imagery noisy label semantic segmentation
DOI	10.1109/TGRS.2021.3068280
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Plan of China[2017YFA0604500] ; National Key Research and Development Plan of China[2017YFB0202204] ; National Key Research and Development Plan of China[2017YFA0604401] ; National Key Research and Development Plan of China[2020YFB0204700] ; National Natural Science Foundation of China[U1839206]
WOS研究方向	Geochemistry & Geophysics ; Engineering ; Remote Sensing ; Imaging Science & Photographic Technology
WOS类目	Geochemistry & Geophysics ; Engineering, Electrical & Electronic ; Remote Sensing ; Imaging Science & Photographic Technology
WOS记录号	WOS:000730619400064
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：15[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/18392
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Fu, Haohuan; Gong, Peng
作者单位	1.Tsinghua Univ, Dept Earth Syst Sci, Beijing 100084, Peoples R China 2.SenseTime Grp Ltd, Beijing 100084, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China 4.Tsinghua Univ, Dept Comp Sci & Technol, Beijing 100084, Peoples R China 5.Chinese Acad Sci, Aerosp Informat Res Inst, State Key Lab Remote Sensing Sci, Beijing 100101, Peoples R China
推荐引用方式 GB/T 7714	Dong, Runmin,Fang, Weizhen,Fu, Haohuan,et al. High-Resolution Land Cover Mapping Through Learning With Noise Correction[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,2022,60:13.
APA	Dong, Runmin,Fang, Weizhen,Fu, Haohuan,Gan, Lin,Wang, Jie,&Gong, Peng.(2022).High-Resolution Land Cover Mapping Through Learning With Noise Correction.IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,60,13.
MLA	Dong, Runmin,et al."High-Resolution Land Cover Mapping Through Learning With Noise Correction".IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 60(2022):13.