A Deep Learning Model for Road Damage Detection After an Earthquake Based on Synthetic Aperture Radar (SAR) and Field Datasets

This article is a new assessment of damaged roads after the Kumamoto earthquake in southern Japan (2016) using remotely sensed synthetic aperture radar (SAR) data, field data and deep learning. Three SAR images from descending orbits of Sentinel-1 in vertical-vertical polarizations are considered fo...

Full description

Saved in:
Bibliographic Details
Published in:IEEE journal of selected topics in applied earth observations and remote sensing 2022-01, Vol.15, p.5753-5765
Main Authors: Karimzadeh, Sadra, Ghasemi, Mohammad, Matsuoka, Masashi, Yagi, Koichi, Zulfikar, Abdullah Can
Format: Article
Language:English
Subjects:
Accelerometers
Damage detection
Datasets
Deep learning
Detection
Earthquake damage
Earthquakes
international roughness index (IRI)
Kumamoto
Machine learning
Multilayer perceptrons
Radar
Radar data
Remote sensing
Roads
Rough surfaces
Roughness
SAR (radar)
Seismic activity
Seismic measurements
Synthetic aperture radar
synthetic aperture radar (SAR)
Transportation
Vertical polarization
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This article is a new assessment of damaged roads after the Kumamoto earthquake in southern Japan (2016) using remotely sensed synthetic aperture radar (SAR) data, field data and deep learning. Three SAR images from descending orbits of Sentinel-1 in vertical-vertical polarizations are considered for radiometric calibration, geocoding and interferometric analyses. Field data in terms of the international roughness index (IRI) were gathered over more than 530 km using a smartphone accelerometer and the BumpRecorder application. The relationship between SAR data and IRI data was investigated in a binary (0 and 1) mode to establish a multilayer perceptron model of damaged and intact roads. We found the remote sensing SAR datasets suitable, not only for the detection of damaged roads, but also as an indicator of road roughness changes. The classification results for damaged and intact roads indicated that our datasets (SAR and field measurements), together with a deep learning model, yielded acceptable overall accuracy (87.1%).
ISSN:1939-1404
2151-1535
DOI:10.1109/JSTARS.2022.3189875