Effective Pan-Sharpening by Multiscale Invertible Neural Network and Heterogeneous Task Distilling

doi:10.1109/TGRS.2022.3199210

	Effective Pan-Sharpening by Multiscale Invertible Neural Network and Heterogeneous Task Distilling
	Zhou, Man 1,3; Huang, Jie 3; Fu, Xueyang 3; Zhao, Feng 3; Hong, Danfeng 2
刊名	IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
	2022
卷号	60
关键词	Heterogeneous knowledge-distilling invertible neural network pan-sharpening transformer
ISSN号	0196-2892
DOI	10.1109/TGRS.2022.3199210
通讯作者	Fu, Xueyang(xyfu@ustc.cdu.cn)
英文摘要	As recognized, the ground-truth multispectral (MS) images possess the complementary information (e.g., high-frequency components) of low-resolution (LR) MS images, which can be considered as privileged information to alleviate the spectral distortion and insufficient spatial texture enhancement. Since existing supervised pan-sharpening methods only utilize the ground-truth MS image to supervise the network training, its potential value has not been fully explored. To accomplish this, we propose a heterogeneous knowledge-distilling pan-sharpening framework that distills pan-sharpening by imitating the ground-truth reconstruction task in both the feature space and network output. In our work, the teacher network performs as a variational autoencoder to extract effective features of the ground-truth MS. The student network, acting as pan-sharpening, is trained by the assistance of the teacher network with the process-oriented feature imitation learning. Moreover, we design a customized information-lossless multiscale invertible neural module to effectively fuse LRMS and panchromatic (PAN) images, producing expected pan-sharpened results. To reduce the artifacts generated by the knowledge distillation process, a knowledge-driven refinement subnetwork is further devised according to the pan-sharpening imaging model. Extensive experimental results on different satellite datasets validate that the proposed network outperforms the state-of-the-art methods both visually and quantitatively. The source code will be released at https://github.com/manman1995/pansharpening .
资助项目	National Natural Science Foundation of China (NSFC)[61901433] ; USTC Research Funds of the Double First-Class Initiative[YD2100002003]
WOS关键词	IMAGE FUSION ; FILTER
WOS研究方向	Geochemistry & Geophysics ; Engineering ; Remote Sensing ; Imaging Science & Photographic Technology
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000856251200002
资助机构	National Natural Science Foundation of China (NSFC) ; USTC Research Funds of the Double First-Class Initiative
内容类型	期刊论文
源URL	[http://ir.hfcas.ac.cn:8080/handle/334002/129147]
专题	中国科学院合肥物质科学研究院
通讯作者	Fu, Xueyang
作者单位	1.Chinese Acad Sci, Hefei Inst Phys Sci, Hefei 230031, Peoples R China 2.Chinese Acad Sci, Aerosp Informat Res Inst, Key Lab Digital Earth Sci, Beijing 100094, Peoples R China 3.Univ Sci & Technol China, Hefei 230026, Peoples R China
推荐引用方式 GB/T 7714	Zhou, Man,Huang, Jie,Fu, Xueyang,et al. Effective Pan-Sharpening by Multiscale Invertible Neural Network and Heterogeneous Task Distilling[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,2022,60.
APA	Zhou, Man,Huang, Jie,Fu, Xueyang,Zhao, Feng,&Hong, Danfeng.(2022).Effective Pan-Sharpening by Multiscale Invertible Neural Network and Heterogeneous Task Distilling.IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,60.
MLA	Zhou, Man,et al."Effective Pan-Sharpening by Multiscale Invertible Neural Network and Heterogeneous Task Distilling".IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 60(2022).