Hyperparameter Configuration Learning for Ship Detection From Synthetic Aperture Radar Images

doi:10.1109/LGRS.2021.3139098

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 遥感图像处理团队

	Hyperparameter Configuration Learning for Ship Detection From Synthetic Aperture Radar Images
	Xu, Nuo 1,2; Huo, Chunlei 1,2; Zhang, Xin 1,2; Cao, Yong 1,2; Pan, Chunhong 1,2
刊名	IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
	2022
卷号	19 页码:5
关键词	Radar polarimetry Synthetic aperture radar Marine vehicles Training Feature extraction Optimization Optical sensors Hyperparameter configuration learning (HCL) object detection reinforcement learning (RL) synthetic aperture radar (SAR)
ISSN号	1545-598X
DOI	10.1109/LGRS.2021.3139098
通讯作者	Huo, Chunlei(clhuo@nlpr.ia.ac.cn)
英文摘要	Detecting ships from synthetic aperture radar (SAR) images is inherently subject to its imaging mechanism. With the development of deep learning, advanced learning-based techniques have been migrated from optical images to SAR images. However, the default hyperparameters (e.g., learning rate, size of the anchor box) predefined by a heuristic strategy on optical images might be suboptimal for SAR datasets. In addition, the low-quality imaging in SAR images further reduces the portability of hyperparameters. To solve this problem, a new optimization method, named reinforcement learning and hyperband (RLH), is proposed to dynamically learn hyperparameter configurations by deep reinforcement learning (DRL), where a neural network is adopted to capture the relationship between different configurations and predict new configurations to further improve the performance. Hyperparameter configuration is able to be automatically learned to accommodate various SAR image datasets, and experiments on two SAR image datasets demonstrate the effectiveness and advantage of the proposed approach.
资助项目	National Key Research and Development Program of China[2018AAA0100400] ; Natural Science Foundation of China[62071466] ; Natural Science Foundation of China[91438105] ; Natural Science Foundation of China[62076242] ; Natural Science Foundation of China[61976208]
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:000742729100003
资助机构	National Key Research and Development Program of China ; Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/47054]
专题	自动化研究所_模式识别国家重点实验室_遥感图像处理团队
通讯作者	Huo, Chunlei
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit NLPR, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Xu, Nuo,Huo, Chunlei,Zhang, Xin,et al. Hyperparameter Configuration Learning for Ship Detection From Synthetic Aperture Radar Images[J]. IEEE GEOSCIENCE AND REMOTE SENSING LETTERS,2022,19:5.
APA	Xu, Nuo,Huo, Chunlei,Zhang, Xin,Cao, Yong,&Pan, Chunhong.(2022).Hyperparameter Configuration Learning for Ship Detection From Synthetic Aperture Radar Images.IEEE GEOSCIENCE AND REMOTE SENSING LETTERS,19,5.
MLA	Xu, Nuo,et al."Hyperparameter Configuration Learning for Ship Detection From Synthetic Aperture Radar Images".IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 19(2022):5.