Deep Few-Shot Learning for Hyperspectral Image Classification

被引:338
作者
Liu, Bing [1 ]
Yu, Xuchu [1 ]
Yu, Anzhu [1 ]
Zhang, Pengqiang [1 ]
Wan, Gang [1 ]
Wang, Ruirui [1 ]
机构
[1] Informat Engn Univ, Inst Surveying & Mapping, Zhengzhou 450001, Henan, Peoples R China
来源
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 2019年 / 57卷 / 04期
基金
中国国家自然科学基金;
关键词
3-D convolutional neural networks (3-D CNNs); few-shot learning; hyperspectral image (HSI) classification; residual learning; SPECTRAL-SPATIAL CLASSIFICATION; SEMI-SUPERVISED CLASSIFICATION; FRAMEWORK; SELECTION; NETWORKS;
D O I
10.1109/TGRS.2018.2872830
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
Deep learning methods have recently been successfully explored for hyperspectral image (HSI) classification. However, training a deep-learning classifier notoriously requires hundreds or thousands of labeled samples. In this paper, a deep few-shot learning method is proposed to address the small sample size problem of HSI classification. There are three novel strategies in the proposed algorithm. First, spectral-spatial features are extracted to reduce the labeling uncertainty via a deep residual 3-D convolutional neural network. Second, the network is trained by episodes to learn a metric space where samples from the same class are close and those from different classes are far. Finally, the testing samples are classified by a nearest neighbor classifier in the learned metric space. The key idea is that the designed network learns a metric space from the training data set. Furthermore, such metric space could generalize to the classes of the testing data set. Note that the classes of the testing data set are not seen in the training data set. Four widely used HSI data sets were used to assess the performance of the proposed algorithm. The experimental results indicate that the proposed method can achieve better classification accuracy than the conventional semisupervised methods with only a few labeled samples.
引用
收藏
页码:2290 / 2304
页数:15
相关论文
共 62 条
[1]  
Ando Rie Kubota, 2007, P 24 INT C MACHINE L, P25
[2]  
[Anonymous], P 3 INT C LEARNING R
[3]  
[Anonymous], J APPL REMOTE SENS
[4]  
[Anonymous], 2013, IEEE T PATTERN ANAL, DOI DOI 10.1109/TPAMI.2012.59
[5]  
[Anonymous], PROC CVPR IEEE
[6]  
[Anonymous], 2017, COMMUN ACM, DOI DOI 10.1145/3065386
[7]  
[Anonymous], 2018, P IEEE C COMP VIS PA
[8]  
[Anonymous], ISPRS J PHOTOGRAM RE
[9]  
[Anonymous], 2016, SAL20160527 U TOK
[10]  
[Anonymous], 1989, IEEE T GEOSCI REMOTE