Next Article in Journal
Tar Spot Disease Quantification Using Unmanned Aircraft Systems (UAS) Data
Next Article in Special Issue
Self-Attention-Based Conditional Variational Auto-Encoder Generative Adversarial Networks for Hyperspectral Classification
Previous Article in Journal
Mangrove Ecosystem Mapping Using Sentinel-1 and Sentinel-2 Satellite Images and Random Forest Algorithm in Google Earth Engine
Previous Article in Special Issue
Unsupervised Haze Removal for High-Resolution Optical Remote-Sensing Images Based on Improved Generative Adversarial Networks
 
 
Article

Remote Sensing Image Scene Classification via Label Augmentation and Intra-Class Constraint

by 1, 1,* and 2,3
1
Harbin Institute of Technology, School of Electronics and Information Engineering, Harbin 150001, China
2
Helmholtz-Zentrum Dresden-Rossendorf, Machine Learning Group, Helmholtz Institute Freiberg for Resource Technology, Chemnitzer Straße 40, 09599 Freiberg, Germany
3
Institute of Advanced Research in Artificial Intelligence (IARAI), Landstraßer Hauptstraße 5, 1030 Vienna, Austria
*
Author to whom correspondence should be addressed.
Academic Editor: Jaime Zabalza
Remote Sens. 2021, 13(13), 2566; https://doi.org/10.3390/rs13132566
Received: 10 May 2021 / Revised: 18 June 2021 / Accepted: 28 June 2021 / Published: 30 June 2021
(This article belongs to the Special Issue Deep Learning for Remote Sensing Data)
In recent years, many convolutional neural network (CNN)-based methods have been proposed to address the scene classification tasks of remote sensing images. Since the number of training samples in RS datasets is generally small, data augmentation is often used to expand the training set. It is, however, not appropriate when original data augmentation methods keep the label and change the content of the image at the same time. In this study, label augmentation (LA) is presented to fully utilize the training set by assigning a joint label to each generated image, which considers the label and data augmentation at the same time. Moreover, the output of images obtained by different data augmentation is aggregated in the test process. However, the augmented samples increase the intra-class diversity of the training set, which is a challenge to complete the following classification process. To address the above issue and further improve classification accuracy, Kullback–Leibler divergence (KL) is used to constrain the output distribution of two training samples with the same scene category to generate a consistent output distribution. Extensive experiments were conducted on widely-used UCM, AID and NWPU datasets. The proposed method can surpass the other state-of-the-art methods in terms of classification accuracy. For example, on the challenging NWPU dataset, competitive overall accuracy (i.e., 91.05%) is obtained with a 10% training ratio. View Full-Text
Keywords: scene classification; remote sensing image; convolutional neural network; label augmentation (LA); joint label; intra-class constraint scene classification; remote sensing image; convolutional neural network; label augmentation (LA); joint label; intra-class constraint
Show Figures

Graphical abstract

MDPI and ACS Style

Xie, H.; Chen, Y.; Ghamisi, P. Remote Sensing Image Scene Classification via Label Augmentation and Intra-Class Constraint. Remote Sens. 2021, 13, 2566. https://doi.org/10.3390/rs13132566

AMA Style

Xie H, Chen Y, Ghamisi P. Remote Sensing Image Scene Classification via Label Augmentation and Intra-Class Constraint. Remote Sensing. 2021; 13(13):2566. https://doi.org/10.3390/rs13132566

Chicago/Turabian Style

Xie, Hao, Yushi Chen, and Pedram Ghamisi. 2021. "Remote Sensing Image Scene Classification via Label Augmentation and Intra-Class Constraint" Remote Sensing 13, no. 13: 2566. https://doi.org/10.3390/rs13132566

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop