Detection of road surface defects from data acquired by a laser scanner

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F21%3APU140776" target="_blank" >RIV/00216305:26220/21:PU140776 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.fekt.vut.cz/conf/EEICT/archiv/sborniky/EEICT_2021_sbornik_2.pdf" target="_blank" >https://www.fekt.vut.cz/conf/EEICT/archiv/sborniky/EEICT_2021_sbornik_2.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Detection of road surface defects from data acquired by a laser scanner

Popis výsledku v původním jazyce

Research in the field of automatic detection of road surface defects has been relatively widespread in recent years. Most of the existing works solve this issue by processing the image acquired by camera technology. The contribution of this study is the proposal of the LRS-CNN algorithm for the detection of defects on road surfaces based on their laser scans. The advantage of LRS-CNN is the ability to detect so-called microcracks, which can not be recognized from camera recordings. We have also found that transfer learning methods are not suitable for the use of road defect detection from their laser scans. Our LRS-CNN algorithm has been trained on unique nonpublic data and is able to achieve up to 99.33 % of success depending on the type of task.

Název v anglickém jazyce

Detection of road surface defects from data acquired by a laser scanner

Popis výsledku anglicky

Research in the field of automatic detection of road surface defects has been relatively widespread in recent years. Most of the existing works solve this issue by processing the image acquired by camera technology. The contribution of this study is the proposal of the LRS-CNN algorithm for the detection of defects on road surfaces based on their laser scans. The advantage of LRS-CNN is the ability to detect so-called microcracks, which can not be recognized from camera recordings. We have also found that transfer learning methods are not suitable for the use of road defect detection from their laser scans. Our LRS-CNN algorithm has been trained on unique nonpublic data and is able to achieve up to 99.33 % of success depending on the type of task.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20203 - Telecommunications

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings II of the 27th Conference STUDENT EEICT 2021

ISBN

978-80-214-5943-4

ISSN

—

e-ISSN

—

Počet stran výsledku

5

Strana od-do

275-279

Název nakladatele

Brno Univeristy of Technology, Faculty of Electrical Engineering and Communication

Místo vydání

Brno, Czech Republic

Místo konání akce

Brno

Datum konání akce

27. 4. 2021

Typ akce podle státní příslušnosti

CST - Celostátní akce

Kód UT WoS článku

—