SegCloud: a novel cloud image segmentation model using a deep convolutional neural network for ground-based all-sky-view camera observation

doi:10.5194/amt-13-1953-2020

HFCAS OpenIR

	SegCloud: a novel cloud image segmentation model using a deep convolutional neural network for ground-based all-sky-view camera observation
	Xie, Wanyi 1,2; Liu, Dong1,2 ; Yang, Ming 4; Chen, Shaoqing 4; Wang, Benge 4; Wang, Zhenzhu1,2 ; Xia, Yingwei3 ; Liu, Yong2,3 ; Wang, Yiren 1,2; Zhang, Chaofang 3
	2020-04-17
发表期刊	ATMOSPHERIC MEASUREMENT TECHNIQUES
ISSN	1867-1381
通讯作者	Wang, Yiren(wyiren90@mail.ustc.edu.cn) ; Zhang, Chaofang(zcf0413@mail.ustc.edu.cn)
摘要	Cloud detection and cloud properties have substantial applications in weather forecast, signal attenuation analysis, and other cloud-related fields. Cloud image segmentation is the fundamental and important step in deriving cloud cover. However, traditional segmentation methods rely on low-level visual features of clouds and often fail to achieve satisfactory performance. Deep convolutional neural networks (CNNs) can extract high-level feature information of objects and have achieved remarkable success in many image segmentation fields. On this basis, a novel deep CNN model named SegCloud is proposed and applied for accurate cloud segmentation based on ground-based observation. Architecturally, SegCloud possesses a symmetric encoder-decoder structure. The encoder network combines low-level cloud features to form high-level, low-resolution cloud feature maps, whereas the decoder network restores the obtained high-level cloud feature maps to the same resolution of input images. The Softmax classifier finally achieves pixel-wise classification and outputs segmentation results. SegCloud has powerful cloud discrimination capability and can automatically segment whole-sky images obtained by a ground-based all-sky-view camera. The performance of SegCloud is validated by extensive experiments, which show that SegCloud is effective and accurate for ground-based cloud segmentation and achieves better results than traditional methods do. The accuracy and practicability of SegCloud are further proven by applying it to cloud cover estimation.
DOI	10.5194/amt-13-1953-2020
关键词[WOS]	CLASSIFICATION ; SYSTEM
收录类别	SCI
语种	英语
资助项目	Science and Technology Service Network Initiative of the Chinese Academy of Sciences[KFJ-STS-QYZD-022] ; Youth Innovation Promotion Association, CAS[2017482] ; Research on Key Technology of Short-Term Forecasting of Photovoltaic Power Generation Based on All-sky Cloud Parameters[201904b11020031]
项目资助者	Science and Technology Service Network Initiative of the Chinese Academy of Sciences ; Youth Innovation Promotion Association, CAS ; Research on Key Technology of Short-Term Forecasting of Photovoltaic Power Generation Based on All-sky Cloud Parameters
WOS研究方向	Meteorology & Atmospheric Sciences
WOS类目	Meteorology & Atmospheric Sciences
WOS记录号	WOS:000527801200002
出版者	COPERNICUS GESELLSCHAFT MBH
引用统计	被引频次：41[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.hfcas.ac.cn:8080/handle/334002/103350
专题	中国科学院合肥物质科学研究院
通讯作者	Wang, Yiren; Zhang, Chaofang
作者单位	1.Chinese Acad Sci, Anhui Inst Opt & Fine Mech, Key Lab Atmospher Opt, Hefei 230088, Peoples R China 2.Univ Sci & Technol China, Grad Sch, Sci Isl Branch, Hefei 230026, Peoples R China 3.Chinese Acad Sci, Anhui Inst Opt & Fine Mech, Optoelect Appl Technol Res Ctr, Hefei 230031, Peoples R China 4.Civil Aviat Adm China, Anhui Air Traff Management Bur, Hefei 230094, Peoples R China
第一作者单位	中科院安徽光学精密机械研究所
通讯作者单位	中科院安徽光学精密机械研究所
推荐引用方式 GB/T 7714	Xie, Wanyi,Liu, Dong,Yang, Ming,et al. SegCloud: a novel cloud image segmentation model using a deep convolutional neural network for ground-based all-sky-view camera observation[J]. ATMOSPHERIC MEASUREMENT TECHNIQUES,2020,13.
APA	Xie, Wanyi.,Liu, Dong.,Yang, Ming.,Chen, Shaoqing.,Wang, Benge.,...&Zhang, Chaofang.(2020).SegCloud: a novel cloud image segmentation model using a deep convolutional neural network for ground-based all-sky-view camera observation.ATMOSPHERIC MEASUREMENT TECHNIQUES,13.
MLA	Xie, Wanyi,et al."SegCloud: a novel cloud image segmentation model using a deep convolutional neural network for ground-based all-sky-view camera observation".ATMOSPHERIC MEASUREMENT TECHNIQUES 13(2020).