Next Article in Journal
A Novel Urban Composition Index Based on Water-Impervious Surface-Pervious Surface (W-I-P) Model for Urban Compositions Mapping Using Landsat Imagery
Next Article in Special Issue
Aerial Imagery Feature Engineering Using Bidirectional Generative Adversarial Networks: A Case Study of the Pilica River Region, Poland
Previous Article in Journal
H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected Network
Previous Article in Special Issue
Intelligent Mapping of Urban Forests from High-Resolution Remotely Sensed Imagery Using Object-Based U-Net-DenseNet-Coupled Network
Open AccessArticle

Post-Disaster Building Damage Detection from Earth Observation Imagery Using Unsupervised and Transferable Anomaly Detecting Generative Adversarial Networks

Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente, 7514 AE Enschede, The Netherlands
*
Author to whom correspondence should be addressed.
Remote Sens. 2020, 12(24), 4193; https://doi.org/10.3390/rs12244193
Received: 13 November 2020 / Revised: 16 December 2020 / Accepted: 16 December 2020 / Published: 21 December 2020
We present an unsupervised deep learning approach for post-disaster building damage detection that can transfer to different typologies of damage or geographical locations. Previous advances in this direction were limited by insufficient qualitative training data. We propose to use a state-of-the-art Anomaly Detecting Generative Adversarial Network (ADGAN) because it only requires pre-event imagery of buildings in their undamaged state. This approach aids the post-disaster response phase because the model can be developed in the pre-event phase and rapidly deployed in the post-event phase. We used the xBD dataset, containing pre- and post- event satellite imagery of several disaster-types, and a custom made Unmanned Aerial Vehicle (UAV) dataset, containing post-earthquake imagery. Results showed that models trained on UAV-imagery were capable of detecting earthquake-induced damage. The best performing model for European locations obtained a recall, precision and F1-score of 0.59, 0.97 and 0.74, respectively. Models trained on satellite imagery were capable of detecting damage on the condition that the training dataset was void of vegetation and shadows. In this manner, the best performing model for (wild)fire events yielded a recall, precision and F1-score of 0.78, 0.99 and 0.87, respectively. Compared to other supervised and/or multi-epoch approaches, our results are encouraging. Moreover, in addition to image classifications, we show how contextual information can be used to create detailed damage maps without the need of a dedicated multi-task deep learning framework. Finally, we formulate practical guidelines to apply this single-epoch and unsupervised method to real-world applications. View Full-Text
Keywords: deep learning; Generative Adversarial Networks; post-disaster; building damage assessment; anomaly detection; Unmanned Aerial Vehicles (UAV); satellite; xBD deep learning; Generative Adversarial Networks; post-disaster; building damage assessment; anomaly detection; Unmanned Aerial Vehicles (UAV); satellite; xBD
Show Figures

Graphical abstract

MDPI and ACS Style

Tilon, S.; Nex, F.; Kerle, N.; Vosselman, G. Post-Disaster Building Damage Detection from Earth Observation Imagery Using Unsupervised and Transferable Anomaly Detecting Generative Adversarial Networks. Remote Sens. 2020, 12, 4193. https://doi.org/10.3390/rs12244193

AMA Style

Tilon S, Nex F, Kerle N, Vosselman G. Post-Disaster Building Damage Detection from Earth Observation Imagery Using Unsupervised and Transferable Anomaly Detecting Generative Adversarial Networks. Remote Sensing. 2020; 12(24):4193. https://doi.org/10.3390/rs12244193

Chicago/Turabian Style

Tilon, Sofia; Nex, Francesco; Kerle, Norman; Vosselman, George. 2020. "Post-Disaster Building Damage Detection from Earth Observation Imagery Using Unsupervised and Transferable Anomaly Detecting Generative Adversarial Networks" Remote Sens. 12, no. 24: 4193. https://doi.org/10.3390/rs12244193

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
Search more from Scilit
 
Search
Back to TopTop