Abstract
Abstract. Degradation and damage detection provides essential information to maintenance workers in routine monitoring and to first responders in post-disaster scenarios. Despite advance in Earth Observation (EO), image analysis and deep learning techniques, the quality and quantity of training data for deep learning is still limited. As a result, no robust method has been found yet that can transfer and generalize well over a variety of geographic locations and typologies of damages. Since damages can be seen as anomalies, occurring sparingly over time and space, we propose to use an anomaly detecting Generative Adversarial Network (GAN) to detect damages. The main advantages of using GANs are that only healthy unannotated images are needed, and that a variety of damages, including the never before seen damage, can be detected. In this study we aimed to investigate 1) the ability of anomaly detecting GANs to detect degradation (potholes and cracks) in asphalt road infrastructures using Mobile Mapper imagery and building damage (collapsed buildings, rubble piles) using post-disaster aerial imagery, and 2) the sensitivity of this method against various types of pre-processing. Our results show that we can detect damages in urban scenes at satisfying levels but not on asphalt roads. Future work will investigate how to further classify the found damages and how to improve damage detection for asphalt roads.
Highlights
Infrastructure and urban services are essential for societies and economies
We argue that our method is the most suitable for damage mapping tasks
These results allow us to make practical suggestions for efficient damage detection in post-disaster scenarios using anomaly detecting GANs: a model should be trained on datasets in which at least shadows are removed
Summary
Infrastructure and urban services are essential for societies and economies. they are increasingly prone to disruptions caused by climate change-induced extreme weather events, rising population numbers and general ageing of structures (Hallegatte et al, 2019). The key to reducing the impact of these disruptions is to increase the resilience of the structures and services In this context, mapping degradation and damage plays an important role. Post-disaster damage mapping aids the assessment of damages, which in turn aids faster post-disaster relief and recovery (Eguchi et al, 2009) For these reasons, damage mapping has been an active field of research for decades. A primary issue is that training datasets are hard to obtain This is because most datasets are tailored to fit specific research areas, which severely limits research to compare methods or to transfer methods to other geographical locations or different typologies of damages (Nex et al, 2019b). Quality and quantity of training data remain important issues. Damages appear in various shapes, sizes or contexts
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
