Combination of LiDAR and image analysis to improve exposure and earthquake vulnerability estimation procedures. Application in Lorca, Spain

We present a procedure for assessing the urban exposure and seismic vulnerability that integrates aerial and satellite images with LiDAR data. It comprises three phases: first, we segment the satellite image to divide the study area into different urban patterns. Second, we extract building footprints and attributes that represent the type of building of each urban pattern. Finally, we assign the seismic vulnerability to each building using two machine learning techniques: logistic regression and Bayesian networks. We apply the procedure in the city of Lorca (SE Spain), where we count on a vulnerability database that we use as ground truth for the validation of results. The outcomes show that both machine learning techniques have similar performance, yielding vulnerability classification results with an accuracy of over 75% (F1 Score) in most of the cases, provided that the building footprints are correctly delineated. The procedure is scalable and can be replicated in different areas as a complement to conventional data gathering approaches. It is especially interesting for disaster risk applications in areas where cadastral databases are inaccessible or field surveys are not viable.