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

SyntEO: Synthetic dataset generation for Earth observation and deep learning - Demonstrated for offshore wind farm detection

Hoeser, Thorsten and Kuenzer, Claudia (2022) SyntEO: Synthetic dataset generation for Earth observation and deep learning - Demonstrated for offshore wind farm detection. ISPRS Journal of Photogrammetry and Remote Sensing, 189, pp. 163-184. Elsevier. doi: 10.1016/j.isprsjprs.2022.04.029. ISSN 0924-2716.

[img] PDF - Published version
22MB

Official URL: https://dx.doi.org/10.1016/j.isprsjprs.2022.04.029

Abstract

With the emergence of deep learning in the last years, new opportunities arose in Earth observation research. Nevertheless, they also brought with them new challenges. The data-hungry training processes of deep learning models demand large, resource expensive, annotated datasets and partly replaced knowledge-driven approaches so that model behaviour and the final prediction process became a black box. The proposed SyntEO approach enables Earth observation researchers to automatically generate large deep learning ready datasets by merging existing and procedural data. SyntEO does this by including expert knowledge in the data generation process in a highly structured manner to control the automatic image and label generation by employing an ontology. In this way, fully controllable experiment environments are set up, which support insights in the model training on the synthetic datasets. Thus, SyntEO makes the learning process approachable, which is an important cornerstone for explainable machine learning. We demonstrate the SyntEO approach by predicting offshore wind farms in Sentinel-1 images on two of the worlds largest offshore wind energy production sites. The largest generated dataset has 90,000 training examples. A basic convolutional neural network for object detection, that is only trained on this synthetic data, confidently detects offshore wind farms by minimising false detections in challenging environments. In addition, four sequential datasets are generated, demonstrating how the SyntEO approach can precisely define the dataset structure and influence the training process. SyntEO is thus a hybrid approach that creates an interface between expert knowledge and data-driven image analysis.

Item URL in elib:https://elib.dlr.de/186736/
Document Type:Article
Title:SyntEO: Synthetic dataset generation for Earth observation and deep learning - Demonstrated for offshore wind farm detection
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Hoeser, Thorstenthorsten.hoeser (at) dlr.dehttps://orcid.org/0000-0002-7179-3664
Kuenzer, Claudiaclaudia.kuenzer (at) dlr.deUNSPECIFIED
Date:17 May 2022
Journal or Publication Title:ISPRS Journal of Photogrammetry and Remote Sensing
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:Yes
Volume:189
DOI :10.1016/j.isprsjprs.2022.04.029
Page Range:pp. 163-184
Publisher:Elsevier
ISSN:0924-2716
Status:Published
Keywords:SyntEO Synthetic training data Explainable machine learning Deep learning CNN Offshore wind farm
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Earth Observation
DLR - Research theme (Project):R - Geoscientific remote sensing and GIS methods, R - Remote Sensing and Geo Research
Location: Oberpfaffenhofen
Institutes and Institutions:German Remote Sensing Data Center
German Remote Sensing Data Center > Land Surface Dynamics
Deposited By: Höser, Thorsten
Deposited On:08 Jun 2022 10:12
Last Modified:23 Jun 2022 08:25

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.