elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Generating Synthetic Sidescan Sonar Snippets Using Transfer-Learning in Generative Adversarial Networks

Steiniger, Yannik and Kraus, Dieter and Meisen, Tobias (2021) Generating Synthetic Sidescan Sonar Snippets Using Transfer-Learning in Generative Adversarial Networks. Journal of Marine Science and Engineering, 9 (3). Multidisciplinary Digital Publishing Institute (MDPI). doi: 10.3390/jmse9030239. ISSN 2077-1312.

[img] PDF - Published version
1MB

Official URL: https://www.mdpi.com/2077-1312/9/3/239

Abstract

The training of a deep learning model requires a large amount of data. In case of sidescan sonar images, the number of snippets from objects of interest is limited. Generative adversarial networks (GAN) have shown to be able to generate photo-realistic images. Hence, we use a GAN to augment a baseline sidescan image dataset with synthetic snippets. Although the training of a GAN with few data samples is likely to cause mode collapse, a combination of pre-training using simple simulated images and fine-tuning with real data reduces this problem. However, for sonar data, we show that this approach of transfer-learning a GAN is sensitive to the pre-training step, meaning that the vanishing of the gradients of the GAN's discriminator becomes a critical problem. Here, we demonstrate how to overcome this problem, and thus how to apply transfer-learning to GANs for generating synthetic sidescan snippets in a more robust way. Additionally, in order to further investigate the GAN's ability to augment a sidescan image dataset, the generated images are analyzed in the image and the frequency domain. The work helps other researchers in the field of sonar image processing to augment their dataset with additional synthetic samples.

Item URL in elib:https://elib.dlr.de/141105/
Document Type:Article
Title:Generating Synthetic Sidescan Sonar Snippets Using Transfer-Learning in Generative Adversarial Networks
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Steiniger, YannikYannik.Steiniger (at) dlr.dehttps://orcid.org/0000-0002-9327-446X
Kraus, DieterCity Univertity of Applied Sciences BremenUNSPECIFIED
Meisen, TobiasUniversity of Wuppertalhttps://orcid.org/0000-0002-1969-559X
Date:24 February 2021
Journal or Publication Title:Journal of Marine Science and Engineering
Refereed publication:Yes
Open Access:Yes
Gold Open Access:Yes
In SCOPUS:Yes
In ISI Web of Science:Yes
Volume:9
DOI :10.3390/jmse9030239
Publisher:Multidisciplinary Digital Publishing Institute (MDPI)
ISSN:2077-1312
Status:Published
Keywords:deep learning; generative adversarial networks; transfer-learning; sidescan sonar; synthetic sonar images
HGF - Research field:other
HGF - Program:other
HGF - Program Themes:other
DLR - Research area:no assignment
DLR - Program:no assignment
DLR - Research theme (Project):no assignment
Location: Bremerhaven
Institutes and Institutions:Institute for the Protection of Maritime Infrastructures > Maritime Security Technologies
Deposited By: Steiniger, Yannik
Deposited On:25 Feb 2021 09:45
Last Modified:25 Feb 2021 09:45

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.