UAV-aided bridge inspection protocol through machine learning with improved visibility images

Euiseok Jeong; Junwon Seo; P. E.James Wacker

doi:10.1016/j.eswa.2022.116791

UAV-aided bridge inspection protocol through machine learning with improved visibility images

Euiseok Jeong, Junwon Seo, P. E.James Wacker

Civil Engineering

Research output: Contribution to journal › Article › peer-review

36 Scopus citations

Abstract

This paper aims to introduce a new bridge inspection protocol using Convolutional Neural Network (CNN)-based machine learning in conjunction with improved visibility images acquired by Unmanned Aerial Vehicles (UAVs). With two UAVs, separate inspections following the proposed protocol were initially performed indoor to quantify the damage state of three concrete columns and four Cross-Laminated Timber (CLT) beams. The protocol using the two UAVs was also adopted to inspect an in-service four-span timber bridge in Pipestone, Minnesota in the United States. During damage identification, various types of visually detected damage were classified through CNN-based machine learning. For image visibility improvement, each image with damage was processed with appropriate adjustment of brightness, contrast, and sharpness to identify and measure the damage in an efficient way. The proposed protocol was found to be capable of bridge damage identification and measurement with an average error of 9.12% when compared to the direct measurements.

Original language	English
Article number	116791
Journal	Expert Systems with Applications
Volume	197
DOIs	https://doi.org/10.1016/j.eswa.2022.116791
State	Published - Jul 1 2022

Bibliographical note

Publisher Copyright:
© 2022 Elsevier Ltd

Funding

Partial financial support for this research was provided by the United States Department of Agriculture (USDA), Forest Service through Joint Venture Agreement No. 18-JV-11111133-031) and in conjunction with the Forest Products Laboratory (FPL). The assistance and cooperation of the Pipestone County engineers is gratefully acknowledged. Partial financial support for this research was provided by the United States Department of Agriculture (USDA), Forest Service through Joint Venture Agreement No. 18-JV-11111133-031 ) and in conjunction with the Forest Products Laboratory (FPL). The assistance and cooperation of the Pipestone County engineers is gratefully acknowledged.

Funders	Funder number
Pipestone County engineers
U.S. Department of Agriculture
U.S. Dept. of Agriculture Forest Service	18-JV-11111133-031
U.S. Dept. of Agriculture Forest Service
USDA Forest Products Laboratory

Keywords

Bridge inspection
Computer vision
Convolutional Neural Network (CNN)
Damage quantification
Drone
Image processing
Machine learning
Visibility improvement

ASJC Scopus subject areas

General Engineering
Computer Science Applications
Artificial Intelligence

Access to Document

10.1016/j.eswa.2022.116791

Cite this

@article{a22114dbdaae404aba0f9044b8676825,

title = "UAV-aided bridge inspection protocol through machine learning with improved visibility images",

abstract = "This paper aims to introduce a new bridge inspection protocol using Convolutional Neural Network (CNN)-based machine learning in conjunction with improved visibility images acquired by Unmanned Aerial Vehicles (UAVs). With two UAVs, separate inspections following the proposed protocol were initially performed indoor to quantify the damage state of three concrete columns and four Cross-Laminated Timber (CLT) beams. The protocol using the two UAVs was also adopted to inspect an in-service four-span timber bridge in Pipestone, Minnesota in the United States. During damage identification, various types of visually detected damage were classified through CNN-based machine learning. For image visibility improvement, each image with damage was processed with appropriate adjustment of brightness, contrast, and sharpness to identify and measure the damage in an efficient way. The proposed protocol was found to be capable of bridge damage identification and measurement with an average error of 9.12\% when compared to the direct measurements.",

keywords = "Bridge inspection, Computer vision, Convolutional Neural Network (CNN), Damage quantification, Drone, Image processing, Machine learning, Visibility improvement",

author = "Euiseok Jeong and Junwon Seo and Wacker, \{P. E.James\}",

note = "Publisher Copyright: {\textcopyright} 2022 Elsevier Ltd",

year = "2022",

month = jul,

day = "1",

doi = "10.1016/j.eswa.2022.116791",

language = "English",

volume = "197",

}

TY - JOUR

T1 - UAV-aided bridge inspection protocol through machine learning with improved visibility images

AU - Jeong, Euiseok

AU - Seo, Junwon

AU - Wacker, P. E.James

PY - 2022/7/1

Y1 - 2022/7/1

N2 - This paper aims to introduce a new bridge inspection protocol using Convolutional Neural Network (CNN)-based machine learning in conjunction with improved visibility images acquired by Unmanned Aerial Vehicles (UAVs). With two UAVs, separate inspections following the proposed protocol were initially performed indoor to quantify the damage state of three concrete columns and four Cross-Laminated Timber (CLT) beams. The protocol using the two UAVs was also adopted to inspect an in-service four-span timber bridge in Pipestone, Minnesota in the United States. During damage identification, various types of visually detected damage were classified through CNN-based machine learning. For image visibility improvement, each image with damage was processed with appropriate adjustment of brightness, contrast, and sharpness to identify and measure the damage in an efficient way. The proposed protocol was found to be capable of bridge damage identification and measurement with an average error of 9.12% when compared to the direct measurements.

AB - This paper aims to introduce a new bridge inspection protocol using Convolutional Neural Network (CNN)-based machine learning in conjunction with improved visibility images acquired by Unmanned Aerial Vehicles (UAVs). With two UAVs, separate inspections following the proposed protocol were initially performed indoor to quantify the damage state of three concrete columns and four Cross-Laminated Timber (CLT) beams. The protocol using the two UAVs was also adopted to inspect an in-service four-span timber bridge in Pipestone, Minnesota in the United States. During damage identification, various types of visually detected damage were classified through CNN-based machine learning. For image visibility improvement, each image with damage was processed with appropriate adjustment of brightness, contrast, and sharpness to identify and measure the damage in an efficient way. The proposed protocol was found to be capable of bridge damage identification and measurement with an average error of 9.12% when compared to the direct measurements.

KW - Bridge inspection

KW - Computer vision

KW - Convolutional Neural Network (CNN)

KW - Damage quantification

KW - Drone

KW - Image processing

KW - Machine learning

KW - Visibility improvement

UR - http://www.scopus.com/inward/record.url?scp=85125529605&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85125529605&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2022.116791

DO - 10.1016/j.eswa.2022.116791

M3 - Article

AN - SCOPUS:85125529605

SN - 0957-4174

VL - 197

JO - Expert Systems with Applications

JF - Expert Systems with Applications

M1 - 116791

ER -

UAV-aided bridge inspection protocol through machine learning with improved visibility images

Abstract

Bibliographical note

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this