A novel 2d-3d cnn with spectral-spatial multi-scale feature fusion for hyperspectral image classification

doi:10.3390/rs13224621

CIOMP OpenIR

	A novel 2d-3d cnn with spectral-spatial multi-scale feature fusion for hyperspectral image classification
	D. Liu; G. Han; P. Liu; H. Yang; X. Sun; Q. Li and J. Wu
	2021
发表期刊	Remote Sensing
ISSN	20724292
卷号	13 期号:22
摘要	Multifarious hyperspectral image (HSI) classification methods based on convolutional neural networks (CNN) have been gradually proposed and achieve a promising classification perfor-mance. However, hyperspectral image classification still suffers from various challenges, including abundant redundant information, insufficient spectral-spatial representation, irregular class distri-bution, and so forth. To address these issues, we propose a novel 2D-3D CNN with spectral-spatial multi-scale feature fusion for hyperspectral image classification, which consists of two feature extraction streams, a feature fusion module as well as a classification scheme. First, we employ two diverse backbone modules for feature representation, that is, the spectral feature and the spatial feature extraction streams. The former utilizes a hierarchical feature extraction module to capture multi-scale spectral features, while the latter extracts multi-stage spatial features by introducing a multi-level fusion structure. With these network units, the category attribute information of HSI can be fully excavated. Then, to output more complete and robust information for classification, a multi-scale spectral-spatial-semantic feature fusion module is presented based on a Decomposition-Reconstruction structure. Last of all, we innovate a classification scheme to lift the classification accuracy. Experimental results on three public datasets demonstrate that the proposed method outperforms the state-of-the-art methods. 2021 by the authors. Licensee MDPI, Basel, Switzerland.
DOI	10.3390/rs13224621
URL	查看原文
收录类别	SCI ; EI
引用统计
文献类型	期刊论文
条目标识符	http://ir.ciomp.ac.cn/handle/181722/65464
专题	中国科学院长春光学精密机械与物理研究所
推荐引用方式 GB/T 7714	D. Liu,G. Han,P. Liu,et al. A novel 2d-3d cnn with spectral-spatial multi-scale feature fusion for hyperspectral image classification[J]. Remote Sensing,2021,13(22).
APA	D. Liu,G. Han,P. Liu,H. Yang,X. Sun,&Q. Li and J. Wu.(2021).A novel 2d-3d cnn with spectral-spatial multi-scale feature fusion for hyperspectral image classification.Remote Sensing,13(22).
MLA	D. Liu,et al."A novel 2d-3d cnn with spectral-spatial multi-scale feature fusion for hyperspectral image classification".Remote Sensing 13.22(2021).

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
A novel 2d-3d cnn wi（7221KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览下载