Attention-Based Pyramid Network for Segmentation and Classification of High-Resolution and Hyperspectral Remote Sensing Images

doi:10.3390/rs12213501

CORC > 成都山地灾害与环境研究所 > 中国科学院水利部成都山地灾害与环境研究所 > 山地灾害与地表过程重点实验室

	Attention-Based Pyramid Network for Segmentation and Classification of High-Resolution and Hyperspectral Remote Sensing Images
	Xu, Qingsong 1,2; Yuan, Xin 3; Ouyang, Chaojun 1,2,4; Zeng, Yue 5
刊名	REMOTE SENSING
	2020
卷号	12 期号:21 页码:3501
关键词	high-resolution and hyperspectral images spatial object distribution diversity spectral information extraction attention-based pyramid network heavy-weight spatial feature fusion pyramid network (FFPNet) spatial-spectral FFPNet
DOI	10.3390/rs12213501
通讯作者	Ouyang, Chaojun(cjouyang@imde.ac.cn)
产权排序	1
文献子类	Article
英文摘要	Unlike conventional natural (RGB) images, the inherent large scale and complex structures of remote sensing images pose major challenges such as spatial object distribution diversity and spectral information extraction when existing models are directly applied for image classification. In this study, we develop an attention-based pyramid network for segmentation and classification of remote sensing datasets. Attention mechanisms are used to develop the following modules: (i) a novel and robust attention-based multi-scale fusion method effectively fuses useful spatial or spectral information at different and same scales; (ii) a region pyramid attention mechanism using region-based attention addresses the target geometric size diversity in large-scale remote sensing images; and (iii) cross-scale attention in our adaptive atrous spatial pyramid pooling network adapts to varied contents in a feature-embedded space. Different forms of feature fusion pyramid frameworks are established by combining these attention-based modules. First, a novel segmentation framework, called the heavy-weight spatial feature fusion pyramid network (FFPNet), is proposed to address the spatial problem of high-resolution remote sensing images. Second, an end-to-end spatial-spectral FFPNet is presented for classifying hyperspectral images. Experiments conducted on ISPRS Vaihingen and ISPRS Potsdam high-resolution datasets demonstrate the competitive segmentation accuracy achieved by the proposed heavy-weight spatial FFPNet. Furthermore, experiments on the Indian Pines and the University of Pavia hyperspectral datasets indicate that the proposed spatial-spectral FFPNet outperforms the current state-of-the-art methods in hyperspectral image classification.
电子版国际标准刊号	2072-4292
资助项目	NSFC[42022054] ; Strategic Priority Research Program of CAS[XDA23090303] ; National Key Research and Development Program of China[2017YFC1501000] ; CAS Youth Innovation Promotion Association
WOS关键词	SEMANTIC SEGMENTATION ; FUSION
WOS研究方向	Remote Sensing
语种	英语
出版者	MDPI
WOS记录号	WOS:000589272800001
资助机构	NSFC ; Strategic Priority Research Program of CAS ; National Key Research and Development Program of China ; CAS Youth Innovation Promotion Association
内容类型	期刊论文
源URL	[http://ir.imde.ac.cn/handle/131551/50737]
专题	成都山地灾害与环境研究所_山地灾害与地表过程重点实验室
通讯作者	Ouyang, Chaojun
作者单位	1.Chinese Acad Sci, Inst Mt Hazards & Environm, Key Lab Mt Hazards & Surface Proc, Chengdu 610041, Peoples R China; 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China; 3.Bell Labs, Murray Hill, NJ 07974 USA; 4.Chinese Acad Sci, CAS Ctr Excellence Tibetan Plateau Earth Sci, Beijing 100101, Peoples R China; 5.Southwest Jiao Tong Univ, Sch Econ & Management, Chengdu 610031, Peoples R China
推荐引用方式 GB/T 7714	Xu, Qingsong,Yuan, Xin,Ouyang, Chaojun,et al. Attention-Based Pyramid Network for Segmentation and Classification of High-Resolution and Hyperspectral Remote Sensing Images[J]. REMOTE SENSING,2020,12(21):3501.
APA	Xu, Qingsong,Yuan, Xin,Ouyang, Chaojun,&Zeng, Yue.(2020).Attention-Based Pyramid Network for Segmentation and Classification of High-Resolution and Hyperspectral Remote Sensing Images.REMOTE SENSING,12(21),3501.
MLA	Xu, Qingsong,et al."Attention-Based Pyramid Network for Segmentation and Classification of High-Resolution and Hyperspectral Remote Sensing Images".REMOTE SENSING 12.21(2020):3501.