Deep Learning Approaches for Crack Detection in Bridge Concrete Structures

Daniel Einarson; Dawit Mengistu

doi:10.1109/ICESIC53714.2022.9783576

Deep Learning Approaches for Crack Detection in Bridge Concrete Structures

Daniel Einarson, Dawit Mengistu

Forskningsoutput: Konferensbidrag › Arbetsdokument (paper) › Peer review

5 Citeringar (Scopus)

Sammanfattning

Deep Learning Approaches for Crack Detection in Bridge Concrete Structure

Originalspråk	Engelska
Sidor	7-12
DOI	https://doi.org/10.1109/ICESIC53714.2022.9783576
Status	Publicerad - 2022
Evenemang	International Conference on Electronic Systems and Intelligent Computing - Chennai, Indien Varaktighet: 2022-apr.-22 → 2022-apr.-23

Konferens

Konferens	International Conference on Electronic Systems and Intelligent Computing
Förkortad titel	ICESIC
Land/Territorium	Indien
Ort	Chennai
Period	22-04-22 → 22-04-23

Nationell ämneskategori

Datavetenskap (10201)

FN:s SDG:er

Detta resultat bidrar till följande hållbara utvecklingsmål:

Åtkomst av dokument

10.1109/ICESIC53714.2022.9783576

5 Citeringar
1 Programvara

Detecting Cracks in Concrete
Bidragets översatta titel: Att upptäcka sprickor i betongEinarson, D. (Utvecklare), 2023
Forskningsoutput: Icke-textbaserad output › Programvara

1 Slutfört

Identifiering av sprickor i betong
Einarson, D. (Forskare) & Mengistu, D. (PI)
22-01-10 → 24-06-28
Projekt: Forskning

Fil

Citera det här

@conference{538dacd9a0db4706b6afe961cccd8651,

title = "Deep Learning Approaches for Crack Detection in Bridge Concrete Structures",

abstract = "Convolutional Neural Networks are among the most effective algorithms for image analysis applications. However, the accuracy of the algorithms depends on the availability of powerful computational resources and the quality of the images used to train the models. This paper investigates ways to build robust models to detect cracks in concrete structures using low resolution images and third-party datasets. Our experiments show that reducing image sizes by a factor of 4 does not significantly impact the accuracy. This is helpful to shorten execution time and hence lower cloud service costs. It is also observed that applying a model trained on one image dataset to detect cracks in images from a different source is not a trivial task.",

author = "Daniel Einarson and Dawit Mengistu",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; International Conference on Electronic Systems and Intelligent Computing, ICESIC ; Conference date: 22-04-2022 Through 23-04-2022",

year = "2022",

doi = "10.1109/ICESIC53714.2022.9783576",

language = "English",

pages = "7--12",

}

TY - CONF

T1 - Deep Learning Approaches for Crack Detection in Bridge Concrete Structures

AU - Einarson, Daniel

AU - Mengistu, Dawit

PY - 2022

Y1 - 2022

N2 - Convolutional Neural Networks are among the most effective algorithms for image analysis applications. However, the accuracy of the algorithms depends on the availability of powerful computational resources and the quality of the images used to train the models. This paper investigates ways to build robust models to detect cracks in concrete structures using low resolution images and third-party datasets. Our experiments show that reducing image sizes by a factor of 4 does not significantly impact the accuracy. This is helpful to shorten execution time and hence lower cloud service costs. It is also observed that applying a model trained on one image dataset to detect cracks in images from a different source is not a trivial task.

AB - Convolutional Neural Networks are among the most effective algorithms for image analysis applications. However, the accuracy of the algorithms depends on the availability of powerful computational resources and the quality of the images used to train the models. This paper investigates ways to build robust models to detect cracks in concrete structures using low resolution images and third-party datasets. Our experiments show that reducing image sizes by a factor of 4 does not significantly impact the accuracy. This is helpful to shorten execution time and hence lower cloud service costs. It is also observed that applying a model trained on one image dataset to detect cracks in images from a different source is not a trivial task.

U2 - 10.1109/ICESIC53714.2022.9783576

DO - 10.1109/ICESIC53714.2022.9783576

M3 - Paper

SP - 7

EP - 12

T2 - International Conference on Electronic Systems and Intelligent Computing

Y2 - 22 April 2022 through 23 April 2022

ER -

Deep Learning Approaches for Crack Detection in Bridge Concrete Structures

Sammanfattning

Konferens

Nationell ämneskategori

FN:s SDG:er

Åtkomst av dokument

Fingeravtryck

Forskningsoutput

Detecting Cracks in Concrete

Projekt

Identifiering av sprickor i betong

Citera det här