Abstract
Remotely sensed data can provide the basis for timely and efficient building damage maps that are of fundamental importance to support the response activities following disaster events. However, the generation of these maps continues to be mainly based on the manual extraction of relevant information in operational frameworks. Considering the identification of visible structural damages caused by earthquakes and explosions, several recent works have shown that Convolutional Neural Networks (CNN) outperform traditional methods. However, the limited availability of publicly available image datasets depicting structural disaster damages, and the wide variety of sensors and spatial resolution used for these acquisitions (from space, aerial and UAV platforms), have limited the clarity of how these networks can effectively serve First Responder needs and emergency mapping service requirements. In this paper, an advanced CNN for visible structural damage detection is tested to shed some light on what deep learning networks can currently deliver, and its adoption in realistic operational conditions after earthquakes and explosions is critically discussed. The heterogeneous and large datasets collected by the authors covering different locations, spatial resolutions and platforms were used to assess the network performances in terms of transfer learning with specific regard to geographical transferability of the trained network to imagery acquired in different locations. The computational time needed to deliver these maps is also assessed. Results show that quality metrics are influenced by the composition of training samples used in the network. To promote their wider use, three pre-trained networks—optimized for satellite, airborne and UAV image spatial resolutions and viewing angles—are made freely available to the scientific community.
Highlights
The localization of damaged buildings in the immediate hours after a disastrous event is one of the first and most important tasks of the emergency response phase [1,2]
In the framework of two EU funded FP7 projects (Reconass [19] and Inachus [20]), several algorithms, mostly based on Convolutional Neural Networks (CNN), have been developed, improving the detection of structural damages caused by earthquakes or explosions [15,16,21,22], which generate comparable damage signatures. Considering these two event typologies, this paper investigates the performances of current CNN for building damage detection in realistic operative conditions, investigating their performance in terms of (i) transfer learning and (ii) running time
Three main aspects were considered during this work: (i) the transfer learning performances considering the geographical transferability of the trained networks, (ii) the improvement given by a soft fine-tuning in the performances of the achieved classification, and (iii) the processing time given an image of defined size
Summary
The localization of damaged buildings in the immediate hours after a disastrous event is one of the first and most important tasks of the emergency response phase [1,2] In this regard, remote sensing is a cost effective and rapid way to inspect the area and exploit the acquired information to organize prompt actions [3]. Optical satellite imagery has been the most widely adopted data source for building damage detection, due to the possibility to image every area of the Earth, with different sensors and spatial resolution, in a few hours/days from the tasking request. Mechanisms and services such as the International Charter “Space and Major Disasters” (IC) and the Copenicus Emergency Management. Manned airplanes are often unavailable in remote areas, especially in the immediate hours after an emergency; in most cases, airborne images to assess the damages and the economic losses only become available several days after the catastrophic event (supporting the cleanup and rehabilitation phases rather than emergency response)
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 callDisclaimer: 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.
