期刊
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
卷 61, 期 -, 页码 -出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TGRS.2023.3277467
关键词
Feature extraction; Convolution; Convolutional neural networks; Data mining; Semantics; Radio frequency; Kernel; Attention; encoder-decoder; global feature; hyperspectral image (HSI) classification; long short view feature
In this article, a dual-view spectral and global spatial feature fusion network (DSGSF) is proposed for hyperspectral image (HSI) classification. The DSGSF consists of a spatial subnetwork and a spectral subnetwork, which can extract spatial-spectral features with strong discriminating performance.
For hyperspectral image (HSI) classification, two branch networks generally use convolutional neural networks (CNNs) to extract the spatial features and long short-term memory (LSTM) to learn the spectral features. However, CNNs with a local kernel neglect the global properties of the whole HSI. LSTM does not consider the macroscopic and detailed information of spectra. In this article, we propose a dual-view spectral and global spatial feature fusion network (DSGSF) to extract the spatial-spectral features for HSI classification (HSIC), including a spatial subnetwork and a spectral subnetwork. In the spatial subnetwork, we propose a global spatial feature representation model based on the encoder-decoder structure with channel attention and spatial attention to learn the global spatial features. In the spectral subnetwork, we design a dual-view spectral feature aggregation model with view attention to learn the diversity of spectral features. By fusing the two subnetworks, we construct DSGSF to extract the spatial-spectral features of HSI with strong discriminating performance. Experimental results on three public datasets illustrate that the proposed method can achieve competitive results compared with the state-of-the-art methods. Code: https://github.com/RZWang-WH/DSGSF.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据