Structural Building Damage Detection with Deep Learning: Assessment of a State-of-the-Art CNN in Operational Conditions

TLDR

Remote‑sensed data can generate timely building‑damage maps essential for disaster response, yet current maps rely mainly on manual extraction; recent CNNs outperform traditional methods, but limited datasets and varied sensor resolutions hinder understanding of their operational effectiveness for first responders. This paper evaluates an advanced CNN for visible structural damage detection to assess its current performance and discuss its adoption in realistic post‑earthquake and post‑explosion operational settings. The authors employed heterogeneous large datasets from multiple locations, spatial resolutions, and platforms to test transfer learning and geographical transferability of the network, also measuring computational time for map delivery. Results indicate that training sample composition influences quality metrics, and the authors released three pre‑trained networks optimized for satellite, airborne, and UAV imagery to promote wider use.

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.