Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-9345
Main Title: Semantic-Fusion Gans for Semi-Supervised Satellite Image Classification
Author(s): Roy, Subhankar
Sangineto, Enver
Demir, Begüm
Sebe, Nicu
Type: Conference Object
Language Code: en
Abstract: Most of the public satellite image datasets contain only a small number of annotated images. The lack of a sufficient quantity of labeled data for training is a bottleneck for the use of modern deep-learning based classification approaches in this domain. In this paper we propose a semi -supervised approach to deal with this problem. We use the discriminator (D) of a Generative Adversarial Network (GAN) as the final classifier, and we train D using both labeled and unlabeled data. The main novelty we introduce is the representation of the visual information fed to D by means of two different channels: the original image and its “semantic” representation, the latter being obtained by means of an external network trained on ImageNet. The two channels are fused in D and jointly used to classify fake images, real labeled and real unlabeled images. We show that using only 100 labeled images, the proposed approach achieves an accuracy close to 69% and a significant improvement with respect to other GAN-based semi-supervised methods. Although we have tested our approach only on satellite images, we do not use any domain-specific knowledge. Thus, our method can be applied to other semi-supervised domains.
URI: https://depositonce.tu-berlin.de/handle/11303/10385
http://dx.doi.org/10.14279/depositonce-9345
Issue Date: 6-Sep-2018
Date Available: 25-Nov-2019
DDC Class: 006 Spezielle Computerverfahren
Subject(s): semi-supervised learning
generative adversarial networks
satellite image classification
Sponsor/Funder: EC/H2020/759764/EU/Accurate and Scalable Processing of Big Data in Earth Observation Fact Sheet/BigEarth
License: http://rightsstatements.org/vocab/InC/1.0/
Proceedings Title: 2018 25th IEEE International Conference on Image Processing (ICIP)
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Publisher Place: New York, NY
Publisher DOI: 10.1109/ICIP.2018.8451836
Page Start: 684
Page End: 688
EISSN: 2381-8549
ISBN: 978-1-4799-7061-2
978-1-4799-7062-9
Notes: © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Appears in Collections:FG Remote Sensing Image Analysis Group » Publications

Files in This Item:
File Description SizeFormat 
roy_etal_2018.pdfAccepted manuscript1.21 MBAdobe PDFThumbnail
View/Open


Items in DepositOnce are protected by copyright, with all rights reserved, unless otherwise indicated.