Local–Global Cross Fusion Network With Gaussian-Initialized Learnable Positional Prompting for Hyperspectral Image Classification

doi:10.1109/TGRS.2023.3335864

	Local–Global Cross Fusion Network With Gaussian-Initialized Learnable Positional Prompting for Hyperspectral Image Classification
	Zhang, Xin 1; Zhang, Rui 1; Li, Ling 2; Li, Wei 1
	2023
发表期刊	IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
ISSN	0196-2892
卷号	61 页码:16
摘要	Deep learning has significantly advanced the field of hyperspectral remote sensing image classification. Among various methods, the classification method based on spectral-spatial features for hyperspectral classification has attracted wide attention because of its exceptional classification performance. However, such methods encounter challenges in handling input sample and feature extraction. Regarding the input sample, current hyperspectral image (HSI) classification methods based on spectral-spatial features treat each pixel of the sample equally, resulting in inadequate attention to valuable pixels within 3-D samples. Regarding feature extraction, the classification methods struggle to effectively extract both local and global information from HSIs. Aiming at solving the above problems, we propose the local-global cross fusion network with Gaussian-initialized positional prompting (LGGNet). LGGNet is designed with an end-to-end architecture, primarily comprising the Gaussian-initialized learnable positional prompting and the local-global cross fusion network. The Gaussian-initialized learnable positional prompting introduces prompting technique into HSI classification, utilizing trainable parameters with prior information to learn the spatial importance of different pixels within a sample for the first time. The local-global cross fusion network combines operations such as 3-D convolutional neural network (CNN) feature extraction, transformer feature extraction, and feature fusion, efficiently integrating local and global features. Extensive experiments showcase that LGGNet achieves state-of-the-art performance with limited training samples on four benchmark datasets, all within a lightweight framework. The relevant code is available at https://github.com/ibelieveican2018/LGGNet.
关键词	3-D convolutional neural network (CNN) hyperspectral image (HSI) classification prompting remote sensing transformer
DOI	10.1109/TGRS.2023.3335864
收录类别	SCI
语种	英语
资助项目	NSF of China
WOS研究方向	Geochemistry & Geophysics ; Engineering ; Remote Sensing ; Imaging Science & Photographic Technology
WOS类目	Geochemistry & Geophysics ; Engineering, Electrical & Electronic ; Remote Sensing ; Imaging Science & Photographic Technology
WOS记录号	WOS:001122847500008
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/38468
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zhang, Rui
作者单位	1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Processors, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Software, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Xin,Zhang, Rui,Li, Ling,等. Local–Global Cross Fusion Network With Gaussian-Initialized Learnable Positional Prompting for Hyperspectral Image Classification[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,2023,61:16.
APA	Zhang, Xin,Zhang, Rui,Li, Ling,&Li, Wei.(2023).Local–Global Cross Fusion Network With Gaussian-Initialized Learnable Positional Prompting for Hyperspectral Image Classification.IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING,61,16.
MLA	Zhang, Xin,et al."Local–Global Cross Fusion Network With Gaussian-Initialized Learnable Positional Prompting for Hyperspectral Image Classification".IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 61(2023):16.