Improving Post-Earthquake Insurance Claim Management:  A Novel Approach to Prioritize Geospatial Data Collection

Pittore, Massimiliano; Wieland, Marc; Errize, Mustafa; Kariptas, Cagatay; Gungor, I

doi:10.3390/ijgi4042401

Cited by 7 publications

(9 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The estimates from the learning machine can be used to guide and iteratively prioritize the manual mapping efforts. An example of a similar approach to prioritize data acquisition in order to improve post-earthquake insurance claim management is given in Pittore et al [40].…”

Section: Discussionmentioning

confidence: 99%

Learning Change from Synthetic Aperture Radar Images: Performance Evaluation of a Support Vector Machine to Detect Earthquake and Tsunami-Induced Changes

Wieland

Yamazaki

2016

Remote Sensing

Self Cite

View full text Add to dashboard Cite

This study evaluates the performance of a Support Vector Machine (SVM) classifier to learn and detect changes in single-and multi-temporal X-and L-band Synthetic Aperture Radar (SAR) images under varying conditions. The purpose is to provide guidance on how to train a powerful learning machine for change detection in SAR images and to contribute to a better understanding of potentials and limitations of supervised change detection approaches. This becomes particularly important on the background of a rapidly growing demand for SAR change detection to support rapid situation awareness in case of natural disasters. The application environment of this study thus focuses on detecting changes caused by the 2011 Tohoku earthquake and tsunami disaster, where single polarized TerraSAR-X and ALOS PALSAR intensity images are used as input. An unprecedented reference dataset of more than 18,000 buildings that have been visually inspected by local authorities for damages after the disaster forms a solid statistical population for the performance experiments. Several critical choices commonly made during the training stage of a learning machine are being assessed for their influence on the change detection performance, including sampling approach, location and number of training samples, classification scheme, change feature space and the acquisition dates of the satellite images. Furthermore, the proposed machine learning approach is compared with the widely used change image thresholding. The study concludes that a well-trained and tuned SVM can provide highly accurate change detections that outperform change image thresholding. While good performance is achieved in the binary change detection case, a distinction between multiple change classes in terms of damage grades leads to poor performance in the tested experimental setting. The major drawback of a machine learning approach is related to the high costs of training. The outcomes of this study, however, indicate that given dynamic parameter tuning, feature selection and an appropriate sampling approach, already small training samples (100 samples per class) are sufficient to produce high change detection rates. Moreover, the experiments show a good generalization ability of SVM which allows transfer and reuse of trained learning machines.

show abstract

Section: Discussionmentioning

confidence: 99%

Learning Change from Synthetic Aperture Radar Images: Performance Evaluation of a Support Vector Machine to Detect Earthquake and Tsunami-Induced Changes

Wieland

Yamazaki

2016

Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…A customized mobile mapping system, designed for seismic exposure and vulnerability characterization (Pittore et al, 2015; Pittore and Wieland, 2013; Wieland et al, 2012), has been employed to rapidly collect geo-referenced omni-directional images of the built environment. Since a complete coverage of the affected areas would be time-consuming and in most cases simply unfeasible due to time and resource constraints, the images are collected following a statistical sampling approach (Pittore et al, 2015). Several different surveys have been conducted in the Kyrgyz Republic and in Tajikistan between 2012 and 2016, for a total of around 7000 buildings remotely surveyed.…”

Section: Application and Resultsmentioning

confidence: 99%

Variable resolution probabilistic modeling of residential exposure and vulnerability for risk applications

2020

Self Cite

View full text Add to dashboard Cite

In risk assessment, the exposure component describes the elements exposed to the natural hazards and susceptible to damage or loss, while the vulnerability component defines the likelihood to incur damage or loss conditional on a given level of hazard intensity. In this article, we propose a novel adaptive approach to exposure modeling which exploits Dirichlet-Multinomial Bayesian updating to implement the incremental assimilation of sparse in situ survey data into probabilistic models described by compositions (proportions). This methodology is complemented by the introduction of a custom spatial aggregation support based on variable-resolution Central Voronoidal Tessellations. The proposed methodology allows for a more consistent integration of empirical observations, typically from engineering surveys, into large-scale models that can also efficiently exploit expert-elicited knowledge. The resulting models are described in a probabilistic framework, and as such allow for a more thorough analysis of the underlying uncertainty. The proposed approach is applied and discussed in five countries in Central Asia.

show abstract

“…These images are then further filtered based on their mutual distance for uniform spatial coverage along the driven path. Because complete coverage of affected areas would be time-consuming and, in most cases, not feasible because of emergency management constraints, the images are collected following a statistical sampling approach (Pittore et al 2015). A stratification scheme was followed to ensure balanced coverage of the damage distribution in the field.…”

Section: Methodology Data Collection and Analysismentioning

confidence: 99%

“…A customized mobile mapping system, originally designed for seismic exposure and vulnerability characterization (Pittore and Wieland 2013, Pittore et al 2015), was employed to rapidly collect georeferenced omnidirectional images of the built environment. The system was composed of an omnidirectional camera with five synchronized optical units, a low-cost global navigation satellite system, and a data collection and processing unit.…”

Section: Methodsmentioning

confidence: 99%

Bayesian Estimation of Macroseismic Intensity from Post-Earthquake Rapid Damage Mapping

et al. 2018

Self Cite

View full text Add to dashboard Cite

The seismological community acknowledges the essential contribution of macroseismic assessment to the compilation of the seismic catalogs used for seismic hazard assessment. Furthermore, macroseismic observations are routinely employed by civil protection authorities in the aftermath of damaging events to improve their decision making. In this paper, we describe a novel methodology for the rapid, probabilistic estimation of macroseismic intensity in the epicentral area of a major event according to the European Macroseismic scale. The methodology includes mobile mapping and a collaborative online platform for rapid post earthquake reconnaissance. A Bayesian scheme is proposed to integrate direct damage observations and prior information, allowing consideration of ancillary data and expert judgment. According to a feasibility study carried out in the area affected by the 2016 Amatrice (Central Italy) earthquake, the proposed methodology should provide a reliable estimation of intensity, efficiently integrating further post earthquake building damage surveys.

show abstract

Improving Post-Earthquake Insurance Claim Management: A Novel Approach to Prioritize Geospatial Data Collection

Cited by 7 publications

References 23 publications

Learning Change from Synthetic Aperture Radar Images: Performance Evaluation of a Support Vector Machine to Detect Earthquake and Tsunami-Induced Changes

Learning Change from Synthetic Aperture Radar Images: Performance Evaluation of a Support Vector Machine to Detect Earthquake and Tsunami-Induced Changes

Variable resolution probabilistic modeling of residential exposure and vulnerability for risk applications

Bayesian Estimation of Macroseismic Intensity from Post-Earthquake Rapid Damage Mapping

Contact Info

Product

Resources

About