Cloud and Cloud Shadow Detection Based on Multiscale 3D-CNN for High Resolution Multispectral Imagery

doi:10.1109/ACCESS.2020.2967590

	Cloud and Cloud Shadow Detection Based on Multiscale 3D-CNN for High Resolution Multispectral Imagery
	Chen, Yang 2; Tang, Luliang 2; Kan, Zihan 2; Latif, Aamir 3; Yang, Xiucheng 4; Bilal, Muhammad 1; Li, Qingquan 2,5
刊名	IEEE ACCESS
	2020
卷号	8 页码:16505-16516
关键词	Cloud detection cloud shadow convolution neural networks multiscale 3D-CNN
ISSN号	2169-3536
DOI	10.1109/ACCESS.2020.2967590
通讯作者	Tang, Luliang(tll@whu.edu.cn)
英文摘要	Cloud and cloud shadow detection is one of the most important tasks for optical remote sensing image preprocessing. It is not an easy task due to the variety and complexity of underlying surfaces, such as the low-albedo objects (water and mountain shadow) and the high-albedo objects (snow and ice). In this study, an end-to-end multiscale 3D-CNN method is proposed for cloud and cloud shadow detection in high resolution multispectral imagery. Specifically, a multiscale learning module is designed to extract cloud and cloud shadow contextual information of different levels. In order to make full use of band information, four band-combination images are inputted into the multiscale 3D-CNN. A joint spectral-spatial information of 3D-convolution layer is developed to fully explore the joint spatial-spectral correlations feature in the input data. Overall, in the experiments undertaken in this paper, the proposed method achieved a mean overall accuracy of 97.27 & x0025; for cloud detection, with a mean precision of 96.02 & x0025; and a mean recall of 95.86 & x0025;. For cloud shadow detection, the proposed method achieved a mean precision of 95.92 & x0025; and a mean recall of 92.86 & x0025;. Experimental results on two validation datasets (GF-1 WFV validation data and ZY-3 validation data) show that the proposed multiscale-3D-CNN method achieved good performance with limited spectral ranges.
资助项目	National Key Research and Development Plan of China[2017YFB0503604] ; National Key Research and Development Plan of China[2016YFE0200400] ; National Natural Science Foundation of China[41971405] ; National Natural Science Foundation of China[41671442] ; National Natural Science Foundation of China[41571430]
WOS关键词	AUTOMATED CLOUD ; NEURAL-NETWORKS ; DEEP
WOS研究方向	Computer Science ; Engineering ; Telecommunications
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000524752200001
资助机构	National Key Research and Development Plan of China ; National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.igsnrr.ac.cn/handle/311030/133758]
专题	中国科学院地理科学与资源研究所
通讯作者	Tang, Luliang
作者单位	1.Nanjing Univ Informat Sci & Technol, Sch Marine Sci, Nanjing 210044, Peoples R China 2.Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan 430079, Peoples R China 3.Univ Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Beijing 10010, Peoples R China 4.Univ Strasbourg, ICube Lab, F-67000 Strasbourg, France 5.Shenzhen Univ, Shenzhen Key Lab Spatial Smart Sensing & Serv, Shenzhen 518060, Peoples R China
推荐引用方式 GB/T 7714	Chen, Yang,Tang, Luliang,Kan, Zihan,et al. Cloud and Cloud Shadow Detection Based on Multiscale 3D-CNN for High Resolution Multispectral Imagery[J]. IEEE ACCESS,2020,8:16505-16516.
APA	Chen, Yang.,Tang, Luliang.,Kan, Zihan.,Latif, Aamir.,Yang, Xiucheng.,...&Li, Qingquan.(2020).Cloud and Cloud Shadow Detection Based on Multiscale 3D-CNN for High Resolution Multispectral Imagery.IEEE ACCESS,8,16505-16516.
MLA	Chen, Yang,et al."Cloud and Cloud Shadow Detection Based on Multiscale 3D-CNN for High Resolution Multispectral Imagery".IEEE ACCESS 8(2020):16505-16516.