Damage detection on building façades using multi-temporal aerial oblique imagery

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

58 Downloads (Pure)

Abstract

Over the past decades, a special interest has been given to remote-sensing imagery to automate the detection of damaged buildings. Given the large areas it may cover and the possibility of automation of the damage detection process, when comparing with lengthy and costly ground observations. Currently, most image-based damage detection approaches rely on Convolutional Neural Networks (CNN). These are used to determine if a given image patch shows damage or not in a binary classification approach. However, such approaches are often trained using image samples containing only debris and rubble piles. Since such approaches often aim at detecting partial or totally collapsed buildings from remote-sensing imagery. Hence, such approaches might not be applicable when the aim is to detect façade damages. This is due to the fact that façade damages also include spalling, cracks and other small signs of damage. Only a few studies focus their damage analysis on the façade and a multi-temporal approach is still missing. In this paper, a multi-temporal approach specifically designed for the image classification of façade damages is presented. To this end, three multi-temporal approaches are compared with two mono-temporal approaches. Regarding the multi-temporal approaches the objective is to understand the optimal fusion between the two imagery epochs within a CNN. The results show that the multi-temporal approaches outperform the mono-temporal ones by up to 22% in accuracy.
Original languageEnglish
Title of host publicationISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Subtitle of host publicationISPRS Geospatial Week 2019
EditorsG. Vosselman, S.J. Oude Elberink, M.Y. Yang
Place of PublicationEnschede
PublisherInternational Society for Photogrammetry and Remote Sensing (ISPRS)
Pages29-36
Number of pages8
VolumeIV
Edition2/W5
DOIs
Publication statusPublished - 29 May 2019
Event4th ISPRS Geospatial Week 2019 - University of Twente, Enschede, Netherlands
Duration: 10 Jun 201914 Jun 2019
Conference number: 4
https://www.gsw2019.org/

Publication series

NameISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
PublisherCopernicus
ISSN (Print)2194-9042

Conference

Conference4th ISPRS Geospatial Week 2019
CountryNetherlands
CityEnschede
Period10/06/1914/06/19
Internet address

Fingerprint

imagery
damage
detection
remote sensing
spalling
image classification
automation
crack
pile

Keywords

  • ITC-GOLD

Cite this

Duarte, D. A., Nex, F., Kerle, N., & Vosselman, G. (2019). Damage detection on building façades using multi-temporal aerial oblique imagery. In G. Vosselman, S. J. Oude Elberink, & M. Y. Yang (Eds.), ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences: ISPRS Geospatial Week 2019 (2/W5 ed., Vol. IV, pp. 29-36). (ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences). Enschede: International Society for Photogrammetry and Remote Sensing (ISPRS). https://doi.org/10.5194/isprs-annals-IV-2-W5-29-2019
Duarte, D.A. ; Nex, F. ; Kerle, N. ; Vosselman, G. / Damage detection on building façades using multi-temporal aerial oblique imagery. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences: ISPRS Geospatial Week 2019. editor / G. Vosselman ; S.J. Oude Elberink ; M.Y. Yang. Vol. IV 2/W5. ed. Enschede : International Society for Photogrammetry and Remote Sensing (ISPRS), 2019. pp. 29-36 (ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences).
@inbook{3042113f8a7d468f81ef68ef6a238dff,
title = "Damage detection on building fa{\cc}ades using multi-temporal aerial oblique imagery",
abstract = "Over the past decades, a special interest has been given to remote-sensing imagery to automate the detection of damaged buildings. Given the large areas it may cover and the possibility of automation of the damage detection process, when comparing with lengthy and costly ground observations. Currently, most image-based damage detection approaches rely on Convolutional Neural Networks (CNN). These are used to determine if a given image patch shows damage or not in a binary classification approach. However, such approaches are often trained using image samples containing only debris and rubble piles. Since such approaches often aim at detecting partial or totally collapsed buildings from remote-sensing imagery. Hence, such approaches might not be applicable when the aim is to detect fa{\cc}ade damages. This is due to the fact that fa{\cc}ade damages also include spalling, cracks and other small signs of damage. Only a few studies focus their damage analysis on the fa{\cc}ade and a multi-temporal approach is still missing. In this paper, a multi-temporal approach specifically designed for the image classification of fa{\cc}ade damages is presented. To this end, three multi-temporal approaches are compared with two mono-temporal approaches. Regarding the multi-temporal approaches the objective is to understand the optimal fusion between the two imagery epochs within a CNN. The results show that the multi-temporal approaches outperform the mono-temporal ones by up to 22{\%} in accuracy.",
keywords = "ITC-GOLD",
author = "D.A. Duarte and F. Nex and N. Kerle and G. Vosselman",
year = "2019",
month = "5",
day = "29",
doi = "10.5194/isprs-annals-IV-2-W5-29-2019",
language = "English",
volume = "IV",
series = "ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences",
publisher = "International Society for Photogrammetry and Remote Sensing (ISPRS)",
pages = "29--36",
editor = "G. Vosselman and {Oude Elberink}, S.J. and M.Y. Yang",
booktitle = "ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences",
edition = "2/W5",

}

Duarte, DA, Nex, F, Kerle, N & Vosselman, G 2019, Damage detection on building façades using multi-temporal aerial oblique imagery. in G Vosselman, SJ Oude Elberink & MY Yang (eds), ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences: ISPRS Geospatial Week 2019. 2/W5 edn, vol. IV, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, International Society for Photogrammetry and Remote Sensing (ISPRS), Enschede, pp. 29-36, 4th ISPRS Geospatial Week 2019, Enschede, Netherlands, 10/06/19. https://doi.org/10.5194/isprs-annals-IV-2-W5-29-2019

Damage detection on building façades using multi-temporal aerial oblique imagery. / Duarte, D.A.; Nex, F.; Kerle, N.; Vosselman, G.

ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences: ISPRS Geospatial Week 2019. ed. / G. Vosselman; S.J. Oude Elberink; M.Y. Yang. Vol. IV 2/W5. ed. Enschede : International Society for Photogrammetry and Remote Sensing (ISPRS), 2019. p. 29-36 (ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences).

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

TY - CHAP

T1 - Damage detection on building façades using multi-temporal aerial oblique imagery

AU - Duarte, D.A.

AU - Nex, F.

AU - Kerle, N.

AU - Vosselman, G.

PY - 2019/5/29

Y1 - 2019/5/29

N2 - Over the past decades, a special interest has been given to remote-sensing imagery to automate the detection of damaged buildings. Given the large areas it may cover and the possibility of automation of the damage detection process, when comparing with lengthy and costly ground observations. Currently, most image-based damage detection approaches rely on Convolutional Neural Networks (CNN). These are used to determine if a given image patch shows damage or not in a binary classification approach. However, such approaches are often trained using image samples containing only debris and rubble piles. Since such approaches often aim at detecting partial or totally collapsed buildings from remote-sensing imagery. Hence, such approaches might not be applicable when the aim is to detect façade damages. This is due to the fact that façade damages also include spalling, cracks and other small signs of damage. Only a few studies focus their damage analysis on the façade and a multi-temporal approach is still missing. In this paper, a multi-temporal approach specifically designed for the image classification of façade damages is presented. To this end, three multi-temporal approaches are compared with two mono-temporal approaches. Regarding the multi-temporal approaches the objective is to understand the optimal fusion between the two imagery epochs within a CNN. The results show that the multi-temporal approaches outperform the mono-temporal ones by up to 22% in accuracy.

AB - Over the past decades, a special interest has been given to remote-sensing imagery to automate the detection of damaged buildings. Given the large areas it may cover and the possibility of automation of the damage detection process, when comparing with lengthy and costly ground observations. Currently, most image-based damage detection approaches rely on Convolutional Neural Networks (CNN). These are used to determine if a given image patch shows damage or not in a binary classification approach. However, such approaches are often trained using image samples containing only debris and rubble piles. Since such approaches often aim at detecting partial or totally collapsed buildings from remote-sensing imagery. Hence, such approaches might not be applicable when the aim is to detect façade damages. This is due to the fact that façade damages also include spalling, cracks and other small signs of damage. Only a few studies focus their damage analysis on the façade and a multi-temporal approach is still missing. In this paper, a multi-temporal approach specifically designed for the image classification of façade damages is presented. To this end, three multi-temporal approaches are compared with two mono-temporal approaches. Regarding the multi-temporal approaches the objective is to understand the optimal fusion between the two imagery epochs within a CNN. The results show that the multi-temporal approaches outperform the mono-temporal ones by up to 22% in accuracy.

KW - ITC-GOLD

UR - https://ezproxy2.utwente.nl/login?url=https://library.itc.utwente.nl/login/2019/chap/duarte_dam.pdf

U2 - 10.5194/isprs-annals-IV-2-W5-29-2019

DO - 10.5194/isprs-annals-IV-2-W5-29-2019

M3 - Chapter

VL - IV

T3 - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

SP - 29

EP - 36

BT - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

A2 - Vosselman, G.

A2 - Oude Elberink, S.J.

A2 - Yang, M.Y.

PB - International Society for Photogrammetry and Remote Sensing (ISPRS)

CY - Enschede

ER -

Duarte DA, Nex F, Kerle N, Vosselman G. Damage detection on building façades using multi-temporal aerial oblique imagery. In Vosselman G, Oude Elberink SJ, Yang MY, editors, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences: ISPRS Geospatial Week 2019. 2/W5 ed. Vol. IV. Enschede: International Society for Photogrammetry and Remote Sensing (ISPRS). 2019. p. 29-36. (ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences). https://doi.org/10.5194/isprs-annals-IV-2-W5-29-2019