Unsupervised Multiview Graph Contrastive Feature Learning for Hyperspectral Image Classification

被引:3
作者
Chang, Yuan [1 ,2 ]
Liu, Quanwei [3 ]
Zhang, Yuxiang [4 ]
Dong, Yanni [2 ]
机构
[1] China Univ Geosci, Sch Geophys & Geomatics, Wuhan 430074, Peoples R China
[2] Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Peoples R China
[3] JamesCook Univ, Coll Sci & Engn, Cairns, Qld 4878, Australia
[4] China Univ Geosci, Sch Geophys & Geomat, Wuhan 430074, Peoples R China
来源
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 2024年 / 62卷
基金
中国国家自然科学基金;
关键词
Feature extraction; Convolution; Classification algorithms; Data augmentation; Hyperspectral imaging; Representation learning; Iron; Contrastive learning (CL); graph convolutional network (GCN); hyperspectral image (HIS) classification; unsupervised feature learning;
D O I
10.1109/TGRS.2024.3431680
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
As a popular deep learning (DL) algorithm, graph neural network (GNN) has been widely used in hyperspectral image (HIS) classification. However, most of the GNN-based classification algorithms are concentrated in the field of semisupervision, which heavily relies on the quantity and quality of samples. To solve this problem, we propose an unsupervised multiview graph contrastive (UMGC) feature learning algorithm to explore the deep semantic features of HSIs without being constrained by samples. First, we construct multiview adjacency matrixes from spatial and spectral directions. Second, the adaptive data augmentation method is used to selectively enhance the topology and attribute structure of the graph. Thereafter, features are extracted by using a contrastive loss to maximize the similarity between the two views. Finally, we tested the model's performance based on multiple evaluation methods. Experimental results on three publicly available hyperspectral datasets show that the proposed UMGC can have better classification performance compared with other state-of-the-art unsupervised feature extraction (FE) methods.
引用
收藏
页数:14
相关论文
共 50 条
[21]   Automatic Graph Learning Convolutional Networks for Hyperspectral Image Classification [J].
Chen, Jie ;
Jiao, Licheng ;
Liu, Xu ;
Li, Lingling ;
Liu, Fang ;
Yang, Shuyuan .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
[22]   Cross-Domain Contrastive Learning for Hyperspectral Image Classification [J].
Guan, Peiyan ;
Lam, Edmund Y. .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
[23]   Domain-Collaborative Contrastive Learning for Hyperspectral Image Classification [J].
Luo, Haiyang ;
Qiao, Xueyi ;
Xu, Yongming ;
Zhong, Shengwei ;
Gong, Chen .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2024, 21
[24]   Superpixel Spectral-Spatial Feature Fusion Graph Convolution Network for Hyperspectral Image Classification [J].
Gong, Zhi ;
Tong, Lei ;
Zhou, Jun ;
Qian, Bin ;
Duan, Lijuan ;
Xiao, Chuangbai .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
[25]   Domain Fusion Contrastive Learning for Cross-Scene Hyperspectral Image Classification [J].
Qiu, Zhao ;
Xu, Jie ;
Peng, Jiangtao ;
Sun, Weiwei .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2025, 63
[26]   Hyperspectral Image Classification Based on Semisupervised Self-Learning and Multiview Information Fusion [J].
Feng, Jia ;
Zhang, Junping ;
Zhang, Ye .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2022, 19
[27]   Unsupervised Discriminative Feature Selection via Contrastive Graph Learning [J].
Zhou, Qian ;
Wang, Qianqian ;
Gao, Quanxue ;
Yang, Ming ;
Gao, Xinbo .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2024, 33 :972-986
[28]   Graph Convolutional Network With Local and Global Feature Fusion for Hyperspectral Image Classification [J].
Wang, Yufan ;
Yu, Xiaodong ;
Dong, Hongbin ;
Zang, Shuying .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2024, 62
[29]   Hyperspectral Imagery Classification Based on Contrastive Learning [J].
Hou, Sikang ;
Shi, Hongye ;
Cao, Xianghai ;
Zhang, Xiaohua ;
Jiao, Licheng .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2022, 60
[30]   Contrastive Graph Convolution Network With Skip Connections for Few-Shot Hyperspectral Image Classification [J].
Liu, Chun ;
Dong, Ao ;
Dong, Dongmei ;
Li, Zheng ;
Yang, Wei ;
Han, Zhigang ;
Guo, Jianzhong .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2024, 21