Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparison
In this paper, we compared five crack detection algorithms using terrestrial laser scanner (TLS) point clouds. The methods are developed based on common point cloud processing knowledge in along- and across-track profiles, surface fitting or local pointwise features, with or without machine learning...
Main Authors: | , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2022-01-01
|
Series: | ISPRS Open Journal of Photogrammetry and Remote Sensing |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2667393221000107 |
_version_ | 1819289034897227776 |
---|---|
author | Ziyi Feng Aimad El Issaoui Matti Lehtomäki Matias Ingman Harri Kaartinen Antero Kukko Joona Savela Hannu Hyyppä Juha Hyyppä |
author_facet | Ziyi Feng Aimad El Issaoui Matti Lehtomäki Matias Ingman Harri Kaartinen Antero Kukko Joona Savela Hannu Hyyppä Juha Hyyppä |
author_sort | Ziyi Feng |
collection | DOAJ |
description | In this paper, we compared five crack detection algorithms using terrestrial laser scanner (TLS) point clouds. The methods are developed based on common point cloud processing knowledge in along- and across-track profiles, surface fitting or local pointwise features, with or without machine learning. The crack area and volume were calculated from the crack points detected by the algorithms. The completeness, correctness, and F1 score of each algorithm were computed against manually collected references. Ten 1-m-by-3.5-m plots containing 75 distresses of six distress types (depression, disintegration, pothole, longitudinal, transverse, and alligator cracks) were selected to explain variability of distresses from a 3-km-long-road. For crack detection at plot level, the best algorithm achieved a completeness of up to 0.844, a correctness of up to 0.853, and an F1 score of up to 0.849. The best algorithm’s overall (ten plots combined) completeness, correctness, and F1 score were 0.642, 0.735, and 0.685 respectively. For the crack area estimation, the overall mean absolute percentage errors (MAPE) of the two best algorithms were 19.8% and 20.3%. In the crack volume estimation, the two best algorithms resulted in 19.3% and 14.5% MAPE. When the plots were grouped based on crack detection complexity, in the ‘easy’ category, the best algorithm reached a crack area estimation MAPE of 8.9%, while for crack volume estimation, the MAPE obtained from the best algorithm was 0.7%. |
first_indexed | 2024-12-24T03:00:26Z |
format | Article |
id | doaj.art-081cf37b73a948579d20c787cc4f919b |
institution | Directory Open Access Journal |
issn | 2667-3932 |
language | English |
last_indexed | 2024-12-24T03:00:26Z |
publishDate | 2022-01-01 |
publisher | Elsevier |
record_format | Article |
series | ISPRS Open Journal of Photogrammetry and Remote Sensing |
spelling | doaj.art-081cf37b73a948579d20c787cc4f919b2022-12-21T17:18:14ZengElsevierISPRS Open Journal of Photogrammetry and Remote Sensing2667-39322022-01-013100010Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparisonZiyi Feng0Aimad El Issaoui1Matti Lehtomäki2Matias Ingman3Harri Kaartinen4Antero Kukko5Joona Savela6Hannu Hyyppä7Juha Hyyppä8Department of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (FGI), the National Land Survey of Finland, Geodeetinrinne 2, 02430, Masala, Finland; Corresponding author.Department of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (FGI), the National Land Survey of Finland, Geodeetinrinne 2, 02430, Masala, FinlandDepartment of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (FGI), the National Land Survey of Finland, Geodeetinrinne 2, 02430, Masala, FinlandDepartment of Built Environment, Aalto University, 02150, Espoo, FinlandDepartment of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (FGI), the National Land Survey of Finland, Geodeetinrinne 2, 02430, Masala, Finland; Department of Geography and Geology, University of Turku, 20500, Turku, FinlandDepartment of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (FGI), the National Land Survey of Finland, Geodeetinrinne 2, 02430, Masala, Finland; Department of Built Environment, Aalto University, 02150, Espoo, FinlandDepartment of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (FGI), the National Land Survey of Finland, Geodeetinrinne 2, 02430, Masala, FinlandDepartment of Built Environment, Aalto University, 02150, Espoo, FinlandDepartment of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (FGI), the National Land Survey of Finland, Geodeetinrinne 2, 02430, Masala, Finland; Department of Built Environment, Aalto University, 02150, Espoo, FinlandIn this paper, we compared five crack detection algorithms using terrestrial laser scanner (TLS) point clouds. The methods are developed based on common point cloud processing knowledge in along- and across-track profiles, surface fitting or local pointwise features, with or without machine learning. The crack area and volume were calculated from the crack points detected by the algorithms. The completeness, correctness, and F1 score of each algorithm were computed against manually collected references. Ten 1-m-by-3.5-m plots containing 75 distresses of six distress types (depression, disintegration, pothole, longitudinal, transverse, and alligator cracks) were selected to explain variability of distresses from a 3-km-long-road. For crack detection at plot level, the best algorithm achieved a completeness of up to 0.844, a correctness of up to 0.853, and an F1 score of up to 0.849. The best algorithm’s overall (ten plots combined) completeness, correctness, and F1 score were 0.642, 0.735, and 0.685 respectively. For the crack area estimation, the overall mean absolute percentage errors (MAPE) of the two best algorithms were 19.8% and 20.3%. In the crack volume estimation, the two best algorithms resulted in 19.3% and 14.5% MAPE. When the plots were grouped based on crack detection complexity, in the ‘easy’ category, the best algorithm reached a crack area estimation MAPE of 8.9%, while for crack volume estimation, the MAPE obtained from the best algorithm was 0.7%.http://www.sciencedirect.com/science/article/pii/S2667393221000107Terrestrial laser scanningPavementRoadCrackDistressPoint cloud |
spellingShingle | Ziyi Feng Aimad El Issaoui Matti Lehtomäki Matias Ingman Harri Kaartinen Antero Kukko Joona Savela Hannu Hyyppä Juha Hyyppä Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparison ISPRS Open Journal of Photogrammetry and Remote Sensing Terrestrial laser scanning Pavement Road Crack Distress Point cloud |
title | Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparison |
title_full | Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparison |
title_fullStr | Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparison |
title_full_unstemmed | Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparison |
title_short | Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparison |
title_sort | pavement distress detection using terrestrial laser scanning point clouds accuracy evaluation and algorithm comparison |
topic | Terrestrial laser scanning Pavement Road Crack Distress Point cloud |
url | http://www.sciencedirect.com/science/article/pii/S2667393221000107 |
work_keys_str_mv | AT ziyifeng pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison AT aimadelissaoui pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison AT mattilehtomaki pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison AT matiasingman pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison AT harrikaartinen pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison AT anterokukko pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison AT joonasavela pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison AT hannuhyyppa pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison AT juhahyyppa pavementdistressdetectionusingterrestriallaserscanningpointcloudsaccuracyevaluationandalgorithmcomparison |