Scene classification for remote sensing images with self-attention augmented CNN

doi:10.1049/ipr2.12540

	Scene classification for remote sensing images with self-attention augmented CNN
	Liu, Zongyin 5; Dong, Anming 3,4; Yu, Jiguo 3,4; Han, Yubing 3,4; Zhou, You 2; Zhao, Kai 1
刊名	IET IMAGE PROCESSING
	2022-05-24
页码	12
ISSN号	1751-9659
DOI	10.1049/ipr2.12540
通讯作者	Dong, Anming(anmingdong@qlu.edu.cn)
英文摘要	Remote sensing scene classification aims to automatically assign a specific semantic label to each image. It is challenging to classify remote sensing scene images due to the images' diversity and rich spatial information. Recently, convolutional neural networks have been widely used to overcome these difficulties, such as the famous Visual Geometry Group (VGG) network. However, the VGG network with local receptive fields cannot model the global information of remote sensing images well. It also needs a large number of parameters and floating point operations to achieve satisfactory accuracy. To overcome these challenges, we introduce the self-attention mechanism to the VGG network. Specifically, we replace the last four convolutional layers in the VGG-19 network with two cascaded self-attention blocks, each consisting of two multi-head self-attention (MHSA) layers with the residual network structure. The new structure can simultaneously explore the local and global information from remote sensing scenes. Such improvements not only reduce model parameters but also improve the classification performance. The effectiveness of the proposed method is validated through experiments on four public data sets, i.e., NaSC-TG2, WHU-RS19, AID and EuroSAT.
资助项目	National Key R&D Program of China[2019YFB2102600] ; National Natural Science Foundation of China[61701269] ; National Natural Science Foundation of China[61832012] ; National Natural Science Foundation of China[61771289] ; Opening Project of Shanghai Trusted Industrial Control Platform[TICPSH202103018-ZC] ; Fundamental Research Enhancement Program of Computer Science and Technology in Qilu University of Technology (Shandong Academy of Sciences)[2021JC02014] ; Joint Research Fund for Young Scholars in Qilu University of Technology (Shandong Academy of Sciences)[2017BSHZ005] ; Program for Youth Innovative Research Team in University of Shandong Province[2019KJN010]
WOS关键词	CONVOLUTIONAL NEURAL-NETWORK ; BENCHMARK
WOS研究方向	Computer Science ; Engineering ; Imaging Science & Photographic Technology
语种	英语
出版者	WILEY
WOS记录号	WOS:000799668900001
资助机构	National Key R&D Program of China ; National Natural Science Foundation of China ; Opening Project of Shanghai Trusted Industrial Control Platform ; Fundamental Research Enhancement Program of Computer Science and Technology in Qilu University of Technology (Shandong Academy of Sciences) ; Joint Research Fund for Young Scholars in Qilu University of Technology (Shandong Academy of Sciences) ; Program for Youth Innovative Research Team in University of Shandong Province
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/49518]
专题	类脑芯片与系统研究
通讯作者	Dong, Anming
作者单位	1.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China 2.Shandong HiCon New Media Inst Co Ltd, Technol Dept, Jinan, Peoples R China 3.Qilu Univ Technol, Sch Math & Stat, Shandong Acad Sci, Jinan, Peoples R China 4.Qilu Univ Technol, Big Data Inst, Shandong Acad Sci, Jinan, Peoples R China 5.Qilu Univ Technol, Sch Comp Sci & Technol, Shandong Acad Sci, Jinan, Peoples R China
推荐引用方式 GB/T 7714	Liu, Zongyin,Dong, Anming,Yu, Jiguo,et al. Scene classification for remote sensing images with self-attention augmented CNN[J]. IET IMAGE PROCESSING,2022:12.
APA	Liu, Zongyin,Dong, Anming,Yu, Jiguo,Han, Yubing,Zhou, You,&Zhao, Kai.(2022).Scene classification for remote sensing images with self-attention augmented CNN.IET IMAGE PROCESSING,12.
MLA	Liu, Zongyin,et al."Scene classification for remote sensing images with self-attention augmented CNN".IET IMAGE PROCESSING (2022):12.