Local spatial information for image super-resolution

doi:10.1016/j.cogsys.2018.06.007

	Local spatial information for image super-resolution
	Zareapoor, Masoumeh 1; Jain, Deepak Kumar 2; Yang, Jie 1
刊名	COGNITIVE SYSTEMS RESEARCH
	2018-12-01
卷号	52 页码:49-57
关键词	Super-resolution Deep convolutional neural network Spatial features Image processing
ISSN号	1389-0417
DOI	10.1016/j.cogsys.2018.06.007
通讯作者	Yang, Jie(jieyang@sjtu.edu.cn)
英文摘要	Image Super resolution plays a crucial role in many applications, such as medical imaging, remote sensing, and security surveillance. Recently convolutional neural network are becoming mainstream in computer vision. Most CNN based super resolution methods cannot fully exploit the entire feature from the original image, and thus the corresponding results will appear low resolution. In this paper, we propose a new network which can reconstruct a high resolution images by upscaling the low resolution images layer by layer with a small scale factor. This strategy helps network to possibly avoid of losing information. The existing CNN models involved bicubic interpolation for preprocessing, which leads to large feature maps and high computational loads. To settle of this problem, the proposed network directly extracts features from the input images, without using preprocessing. In addition, the proposed network investigates the spatial information which is represented by dissimilarities between a low resolution image and its corresponding high resolution by adopting a global residual learning. This differentiable strategy is inserted into the proposed network, to dynamically extract the feature maps. The proposed model not only achieves a compatible performance with the existing prominent methods but also, efficiently reduce the computational expenses. (C) 2018 Elsevier B.V. All rights reserved.
资助项目	NSFC, China[61572315] ; Committee of Science and Technology, Shanghai, China[17JC1403000]
WOS研究方向	Computer Science ; Neurosciences & Neurology ; Psychology
语种	英语
出版者	ELSEVIER SCIENCE BV
WOS记录号	WOS:000450854400006
资助机构	NSFC, China ; Committee of Science and Technology, Shanghai, China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/25715]
专题	中国科学院自动化研究所
通讯作者	Yang, Jie
作者单位	1.Shanghai Jiao Tong Univ, Inst Image Proc & Pattern Recognit, Shanghai 200240, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Zareapoor, Masoumeh,Jain, Deepak Kumar,Yang, Jie. Local spatial information for image super-resolution[J]. COGNITIVE SYSTEMS RESEARCH,2018,52:49-57.
APA	Zareapoor, Masoumeh,Jain, Deepak Kumar,&Yang, Jie.(2018).Local spatial information for image super-resolution.COGNITIVE SYSTEMS RESEARCH,52,49-57.
MLA	Zareapoor, Masoumeh,et al."Local spatial information for image super-resolution".COGNITIVE SYSTEMS RESEARCH 52(2018):49-57.