Contextual classification using photometry and elevation data for damage detection after an earthquake event

Ewelina Rupnik (Corresponding Author), Francesco Nex, Isabella Toschi, Fabio Remondino

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Scopus)
7 Downloads (Pure)

Abstract

This research presents a processing workflow to automatically find damaged building areas in an urban context. The input data requirements are high-resolution multi-view images, acquired from airborne platform. The elevations are derived from a dense surface model generated with photogrammetric methods. With the principal objective of rapid response in emergency situations, two different processing roadmaps are proposed, semi-supervised and unsupervised. Both of them follow a two-step workflow of building detection and building health estimation. Optionally, cadastral layers may serve as a-priori knowledge on building location. The semi-supervised approach involves a data training step, while the unsupervised approach exploits the similarities and dissimilarities between sets of features calculated over the detected buildings. The change detection task is formulated as a classification task defined over a conditional random field. The algorithms are evaluated using two datasets (Vexcel and Midas cameras) and results are compared with ground truth data and specific metrics.
Original languageEnglish
Pages (from-to)543-557
Number of pages15
JournalEuropean journal of remote sensing
Volume51
Issue number1
DOIs
Publication statusPublished - 31 May 2018

Fingerprint

Photometry
earthquake event
Damage Detection
Damage detection
Earthquake
Earthquakes
damage
Work Flow
Processing
Conditional Random Fields
Change Detection
Cameras
Health
Dissimilarity
Emergency
High Resolution
Camera
Metric
Requirements
detection

Keywords

  • ITC-ISI-JOURNAL-ARTICLE
  • ITC-GOLD

Cite this

@article{d68a6f588c06420ca2e27582a91174ee,
title = "Contextual classification using photometry and elevation data for damage detection after an earthquake event",
abstract = "This research presents a processing workflow to automatically find damaged building areas in an urban context. The input data requirements are high-resolution multi-view images, acquired from airborne platform. The elevations are derived from a dense surface model generated with photogrammetric methods. With the principal objective of rapid response in emergency situations, two different processing roadmaps are proposed, semi-supervised and unsupervised. Both of them follow a two-step workflow of building detection and building health estimation. Optionally, cadastral layers may serve as a-priori knowledge on building location. The semi-supervised approach involves a data training step, while the unsupervised approach exploits the similarities and dissimilarities between sets of features calculated over the detected buildings. The change detection task is formulated as a classification task defined over a conditional random field. The algorithms are evaluated using two datasets (Vexcel and Midas cameras) and results are compared with ground truth data and specific metrics.",
keywords = "ITC-ISI-JOURNAL-ARTICLE, ITC-GOLD",
author = "Ewelina Rupnik and Francesco Nex and Isabella Toschi and Fabio Remondino",
year = "2018",
month = "5",
day = "31",
doi = "10.1080/22797254.2018.1458584",
language = "English",
volume = "51",
pages = "543--557",
journal = "European journal of remote sensing",
issn = "2279-7254",
publisher = "Taylor & Francis",
number = "1",

}

Contextual classification using photometry and elevation data for damage detection after an earthquake event. / Rupnik, Ewelina (Corresponding Author); Nex, Francesco; Toschi, Isabella; Remondino, Fabio.

In: European journal of remote sensing, Vol. 51, No. 1, 31.05.2018, p. 543-557.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Contextual classification using photometry and elevation data for damage detection after an earthquake event

AU - Rupnik, Ewelina

AU - Nex, Francesco

AU - Toschi, Isabella

AU - Remondino, Fabio

PY - 2018/5/31

Y1 - 2018/5/31

N2 - This research presents a processing workflow to automatically find damaged building areas in an urban context. The input data requirements are high-resolution multi-view images, acquired from airborne platform. The elevations are derived from a dense surface model generated with photogrammetric methods. With the principal objective of rapid response in emergency situations, two different processing roadmaps are proposed, semi-supervised and unsupervised. Both of them follow a two-step workflow of building detection and building health estimation. Optionally, cadastral layers may serve as a-priori knowledge on building location. The semi-supervised approach involves a data training step, while the unsupervised approach exploits the similarities and dissimilarities between sets of features calculated over the detected buildings. The change detection task is formulated as a classification task defined over a conditional random field. The algorithms are evaluated using two datasets (Vexcel and Midas cameras) and results are compared with ground truth data and specific metrics.

AB - This research presents a processing workflow to automatically find damaged building areas in an urban context. The input data requirements are high-resolution multi-view images, acquired from airborne platform. The elevations are derived from a dense surface model generated with photogrammetric methods. With the principal objective of rapid response in emergency situations, two different processing roadmaps are proposed, semi-supervised and unsupervised. Both of them follow a two-step workflow of building detection and building health estimation. Optionally, cadastral layers may serve as a-priori knowledge on building location. The semi-supervised approach involves a data training step, while the unsupervised approach exploits the similarities and dissimilarities between sets of features calculated over the detected buildings. The change detection task is formulated as a classification task defined over a conditional random field. The algorithms are evaluated using two datasets (Vexcel and Midas cameras) and results are compared with ground truth data and specific metrics.

KW - ITC-ISI-JOURNAL-ARTICLE

KW - ITC-GOLD

UR - https://ezproxy2.utwente.nl/login?url=https://webapps.itc.utwente.nl/library/2018/isi/nex_con.pdf

U2 - 10.1080/22797254.2018.1458584

DO - 10.1080/22797254.2018.1458584

M3 - Article

VL - 51

SP - 543

EP - 557

JO - European journal of remote sensing

JF - European journal of remote sensing

SN - 2279-7254

IS - 1

ER -