KrakN: Transfer Learning framework and dataset for infrastructure thin crack detection

Monitoring the technical condition of infrastructure is a crucial element of its maintenance. Although there are many deep learning models intended for this purpose, they are severely limited in their application due to labour-intensive gathering of new datasets and high demand for computing power d...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Mateusz Żarski, Bartosz Wójcik, Jarosław Adam Miszczak
Formato:	article
Lenguaje:	EN
Publicado:	Elsevier 2021
Materias:	Infrastructure maintenance Structural health monitoring Deep learning Transfer learning Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/066bca083ea24379b22933f5afbdc6f9
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:066bca083ea24379b22933f5afbdc6f9
record_format	dspace
spelling	oai:doaj.org-article:066bca083ea24379b22933f5afbdc6f92021-11-28T04:34:37ZKrakN: Transfer Learning framework and dataset for infrastructure thin crack detection2352-711010.1016/j.softx.2021.100893https://doaj.org/article/066bca083ea24379b22933f5afbdc6f92021-12-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2352711021001503https://doaj.org/toc/2352-7110Monitoring the technical condition of infrastructure is a crucial element of its maintenance. Although there are many deep learning models intended for this purpose, they are severely limited in their application due to labour-intensive gathering of new datasets and high demand for computing power during model training. To overcome these limiting factors we propose a KrakN framework. It enables end-to-end development of unique infrastructure defect detectors on digital images, achieving an accuracy of above 90%. The framework also supports the semi-automatic creation of new datasets and has modest computing power requirements. It can be used to immediately implement deep learning in the process of infrastructure management and, due to its architecture based on transfer learning, allows low metric loss when used across different datasets. We also demonstrate that thanks to its scalability and modular structure, the presented framework is easily modifiable, and can be used in realistic scenarios.Mateusz ŻarskiBartosz WójcikJarosław Adam MiszczakElsevierarticleInfrastructure maintenanceStructural health monitoringDeep learningTransfer learningComputer softwareQA76.75-76.765ENSoftwareX, Vol 16, Iss , Pp 100893- (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Infrastructure maintenance Structural health monitoring Deep learning Transfer learning Computer software QA76.75-76.765
spellingShingle	Infrastructure maintenance Structural health monitoring Deep learning Transfer learning Computer software QA76.75-76.765 Mateusz Żarski Bartosz Wójcik Jarosław Adam Miszczak KrakN: Transfer Learning framework and dataset for infrastructure thin crack detection
description	Monitoring the technical condition of infrastructure is a crucial element of its maintenance. Although there are many deep learning models intended for this purpose, they are severely limited in their application due to labour-intensive gathering of new datasets and high demand for computing power during model training. To overcome these limiting factors we propose a KrakN framework. It enables end-to-end development of unique infrastructure defect detectors on digital images, achieving an accuracy of above 90%. The framework also supports the semi-automatic creation of new datasets and has modest computing power requirements. It can be used to immediately implement deep learning in the process of infrastructure management and, due to its architecture based on transfer learning, allows low metric loss when used across different datasets. We also demonstrate that thanks to its scalability and modular structure, the presented framework is easily modifiable, and can be used in realistic scenarios.
format	article
author	Mateusz Żarski Bartosz Wójcik Jarosław Adam Miszczak
author_facet	Mateusz Żarski Bartosz Wójcik Jarosław Adam Miszczak
author_sort	Mateusz Żarski
title	KrakN: Transfer Learning framework and dataset for infrastructure thin crack detection
title_short	KrakN: Transfer Learning framework and dataset for infrastructure thin crack detection
title_full	KrakN: Transfer Learning framework and dataset for infrastructure thin crack detection
title_fullStr	KrakN: Transfer Learning framework and dataset for infrastructure thin crack detection
title_full_unstemmed	KrakN: Transfer Learning framework and dataset for infrastructure thin crack detection
title_sort	krakn: transfer learning framework and dataset for infrastructure thin crack detection
publisher	Elsevier
publishDate	2021
url	https://doaj.org/article/066bca083ea24379b22933f5afbdc6f9
work_keys_str_mv	AT mateuszzarski krakntransferlearningframeworkanddatasetforinfrastructurethincrackdetection AT bartoszwojcik krakntransferlearningframeworkanddatasetforinfrastructurethincrackdetection AT jarosławadammiszczak krakntransferlearningframeworkanddatasetforinfrastructurethincrackdetection
_version_	1718408299155554304

KrakN: Transfer Learning framework and dataset for infrastructure thin crack detection

Ejemplares similares