Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial Images
A methodology for the automated identification of building damage from post-disaster aerial images was developed based on convolutional neural network (CNN) and building damage inventories. The aerial images and the building damage data obtained in the 2016 Kumamoto, and the 1995 Kobe, Japan earthqu...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/12/1924 |
| _version_ | 1797565361639391232 |
|---|---|
| author | Hiroyuki Miura Tomohiro Aridome Masashi Matsuoka |
| author_facet | Hiroyuki Miura Tomohiro Aridome Masashi Matsuoka |
| author_sort | Hiroyuki Miura |
| collection | DOAJ |
| description | A methodology for the automated identification of building damage from post-disaster aerial images was developed based on convolutional neural network (CNN) and building damage inventories. The aerial images and the building damage data obtained in the 2016 Kumamoto, and the 1995 Kobe, Japan earthquakes were analyzed. Since the roofs of many moderately damaged houses are covered with blue tarps immediately after disasters, not only collapsed and non-collapsed buildings but also the buildings covered with blue tarps were identified by the proposed method. The CNN architecture developed in this study correctly classifies the building damage with the accuracy of approximately 95 % in both earthquake data. We applied the developed CNN model to aerial images in Chiba, Japan, damaged by the typhoon in September 2019. The result shows that more than 90 % of the building damage are correctly classified by the CNN model. |
| first_indexed | 2024-03-10T19:11:04Z |
| format | Article |
| id | doaj.art-5a14b8327bfd49ab8f3df618cf5a8d99 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T19:11:04Z |
| publishDate | 2020-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-5a14b8327bfd49ab8f3df618cf5a8d992023-11-20T03:46:56ZengMDPI AGRemote Sensing2072-42922020-06-011212192410.3390/rs12121924Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial ImagesHiroyuki Miura0Tomohiro Aridome1Masashi Matsuoka2Department of Advanced Science and Engineering, Hiroshima University, Kagamiyama 1-4-1, Higashi-Hiroshima, Hiroshima 739-8527, JapanDepartment of Architecture, Hiroshima University, Kagamiyama 1-4-1, Higashi-Hiroshima, Hiroshima 739-8527, JapanDepartment of Architecture and Building Engineering, Tokyo Institute of Technology, Nagatsuta 4259, Yokohama, Kanagawa 226-8502, JapanA methodology for the automated identification of building damage from post-disaster aerial images was developed based on convolutional neural network (CNN) and building damage inventories. The aerial images and the building damage data obtained in the 2016 Kumamoto, and the 1995 Kobe, Japan earthquakes were analyzed. Since the roofs of many moderately damaged houses are covered with blue tarps immediately after disasters, not only collapsed and non-collapsed buildings but also the buildings covered with blue tarps were identified by the proposed method. The CNN architecture developed in this study correctly classifies the building damage with the accuracy of approximately 95 % in both earthquake data. We applied the developed CNN model to aerial images in Chiba, Japan, damaged by the typhoon in September 2019. The result shows that more than 90 % of the building damage are correctly classified by the CNN model.https://www.mdpi.com/2072-4292/12/12/1924deep learningbuilding damageaerial imageearthquaketyphoon |
| spellingShingle | Hiroyuki Miura Tomohiro Aridome Masashi Matsuoka Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial Images Remote Sensing deep learning building damage aerial image earthquake typhoon |
| title | Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial Images |
| title_full | Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial Images |
| title_fullStr | Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial Images |
| title_full_unstemmed | Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial Images |
| title_short | Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial Images |
| title_sort | deep learning based identification of collapsed non collapsed and blue tarp covered buildings from post disaster aerial images |
| topic | deep learning building damage aerial image earthquake typhoon |
| url | https://www.mdpi.com/2072-4292/12/12/1924 |
| work_keys_str_mv | AT hiroyukimiura deeplearningbasedidentificationofcollapsednoncollapsedandbluetarpcoveredbuildingsfrompostdisasteraerialimages AT tomohiroaridome deeplearningbasedidentificationofcollapsednoncollapsedandbluetarpcoveredbuildingsfrompostdisasteraerialimages AT masashimatsuoka deeplearningbasedidentificationofcollapsednoncollapsedandbluetarpcoveredbuildingsfrompostdisasteraerialimages |