Semisupervised Classification for Hyperspectral Images Using Graph Attention Networks

Semisupervised Classification for Hyperspectral Images Using Graph Attention Networks
复制标题

使用图注意网络对高光谱图像进行半监督分类

DOI:
10.1109/lgrs.2020.2966239
复制
发表时间:
2021-01
影响因子:
4.8
通讯作者:
Liming Zhang
Liming Zhang
中科院分区:
工程技术2区
文献类型:
--
作者:
Anshu Sha;Bin Wang;Xiaofeng Wu;Liming Zhang

文献摘要

参考文献

相似文献

对于高光谱图像,高维数据和有限的标记样本之间的不平衡一直是分类任务的主要障碍。作为一种解决方案,利用标记和未标记样本的半监督学习已经显示出其潜力。在这封信中,提出了一种新的基于图注意力网络(GATs)的HSI半监督分类框架。为充分利用空间信息,设计了空间-光谱联合测量的图形模型构建方法。在卷积过程中,根据相邻节点的注意力系数为不同的节点分配不同的权值,避免了以往图卷积网络(GCN)中人为设计连接权值的问题。在不同背景和分辨率的多个高光谱数据集上的实验结果表明,该方法优于几种最先进的基于图的方法。
For hyperspectral images (HSIs), the imbalance between the high dimensionality and the limited labeled samples has been a main obstacle to classification task. As a solution, semisupervised learning utilizing both labeled and unlabeled samples has shown its potential. In this letter, a novel semisupervised classification framework based on graph attention networks (GATs) for HSIs is proposed. Spatial–spectral joint measurement is designed for the graph model construction to make full use of spatial information. In the convolution process, different weights are assigned to different neighboring nodes according to their attention coefficients, avoiding designing connection weights artificially in previous graph convolution networks (GCNs). Experimental results on multiple hyperspectral data sets with various contexts and resolutions demonstrate that the proposed method outperforms several state-of-the-art graph-based methods.
DOI: 10.1109/lgrs.2008.916070
发表时间: 2008-07-01
影响因子: 4.8
作者:
Gomez-Chova, Luis;Camps-Valls, Gustavo;Calpe, Javier
通讯作者: Calpe, Javier
DOI: 10.1109/msp.2013.2279179
发表时间: 2014-01-01
影响因子: 14.9
作者:
Camps-Valls, Gustavo;Tuia, Devis;Benediktsson, Jon Atli
通讯作者: Benediktsson, Jon Atli
DOI: 10.1016/j.isprsjprs.2014.08.016
发表时间: 2014-11
影响因子: 12.7
作者:
Liguo Wang;Siyuan Hao;Qunming Wang;Ying Wang
通讯作者: Liguo Wang;Siyuan Hao;Qunming Wang;Ying Wang
DOI: 10.17863/cam.48429
发表时间: 2017-10
期刊: ArXiv
影响因子: --
作者:
Petar Velickovic;Guillem Cucurull;Arantxa Casanova;Adriana Romero;P. Lio’;Yoshua Bengio
通讯作者: Petar Velickovic;Guillem Cucurull;Arantxa Casanova;Adriana Romero;P. Lio’;Yoshua Bengio
R-VCANet:一种新的基于深度学习的高光谱图像分类方法
DOI: 10.1109/jstars.2017.2655516
发表时间: 2017-02
影响因子: 5.5
作者:
Pan Bin;Shi Zhenwei;Xu Xia
通讯作者: Xu Xia