No Cover Image

Journal article 311 views 56 downloads

Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition

Hui Yao, Yanhao Liu, Haotian Lv, Ju Huyan, Zhanping You, Yue Hou Orcid Logo

Computer-Aided Civil and Infrastructure Engineering, Volume: 39, Issue: 10, Pages: 1490 - 1506

Swansea University Author: Yue Hou Orcid Logo

  • 65225 VOR Yue HOU.pdf

    PDF | Version of Record

    © 2023 The Authors. Distributed under the terms of a Creative Commons Attribution 4.0 International License (CC BY 4.0).

    Download (5.99MB)

Check full text

DOI (Published version): 10.1111/mice.13128

Abstract

Timely and accurate extraction of pavement crack information is crucial to maintain service conditions and structural safety for infrastructures and reduce further road maintenance costs. Currently, deep learning techniques for automated pavement crack detection are far superior to traditional manua...

Full description

Published in: Computer-Aided Civil and Infrastructure Engineering
ISSN: 1093-9687 1467-8667
Published: Wiley 2024
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa65225
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2023-12-06T11:47:06Z
last_indexed 2023-12-06T11:47:06Z
id cronfa65225
recordtype SURis
fullrecord <?xml version="1.0" encoding="utf-8"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"><bib-version>v2</bib-version><id>65225</id><entry>2023-12-06</entry><title>Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition</title><swanseaauthors><author><sid>92bf566c65343cb3ee04ad963eacf31b</sid><ORCID>0000-0002-4334-2620</ORCID><firstname>Yue</firstname><surname>Hou</surname><name>Yue Hou</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2023-12-06</date><deptcode>ACEM</deptcode><abstract>Timely and accurate extraction of pavement crack information is crucial to maintain service conditions and structural safety for infrastructures and reduce further road maintenance costs. Currently, deep learning techniques for automated pavement crack detection are far superior to traditional manual approaches in both speed and accuracy. However, existing deep learning models may easily lose crack details when processing images containing complex background textures or other noises. Although many studies have alleviated this challenge by introducing attention mechanisms, especially the non-local (NL) block, which has the ability to efficiently capture long-range dependencies to facilitate crack pixel capture, the huge computational cost of NL makes the inference time of the model too long, which is not conducive to practical implementation. In this study, a new module, namely, the pyramid region attention module (PRAM), was developed by combining the pyramid pooling module in the pyramid scene parsing network and optimized NL, which can achieve global multi-scale context integration and long-range dependencies capture at a relatively lower computational cost. By applying PRAM to deep skip connections in the modified U-Net, an effective crack segmentation model called CrackResU-Net was developed. The test results on the existing CrackForest dataset showed that CrackResU-Net not only achieved an F1 score of 0.9580 but also took only 25.89 ms to process an image with a resolution of 480 × 320, which had advantages in accuracy and speed, compared with several other state-of-the-art crack segmentation approaches. It was fully demonstrated that this approach could realize automatic fast and high-precision recognition of pavement cracks for engineering purposes.</abstract><type>Journal Article</type><journal>Computer-Aided Civil and Infrastructure Engineering</journal><volume>39</volume><journalNumber>10</journalNumber><paginationStart>1490</paginationStart><paginationEnd>1506</paginationEnd><publisher>Wiley</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>1093-9687</issnPrint><issnElectronic>1467-8667</issnElectronic><keywords/><publishedDay>15</publishedDay><publishedMonth>5</publishedMonth><publishedYear>2024</publishedYear><publishedDate>2024-05-15</publishedDate><doi>10.1111/mice.13128</doi><url/><notes/><college>COLLEGE NANME</college><department>Aerospace, Civil, Electrical, and Mechanical Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>ACEM</DepartmentCode><institution>Swansea University</institution><apcterm>SU Library paid the OA fee (TA Institutional Deal)</apcterm><funders>The authors appreciate the financial support from Hunan Expressway Group Co. Ltd and the Hunan Department of Transportation (No. 202152) in China. The authors also appreciate the funding support from the National Natural Science Foundation of China (No. 51778038), the China National Key R&amp;D Program (2021YFB2600600, 2021YFB2600604, 2018YFB1600202), National Natural Science Foundation of China (grant number: 52108403).</funders><projectreference>No. 202152, No. 51778038, 2021YFB2600600, 2021YFB2600604, 2018YFB1600202, grant number: 52108403</projectreference><lastEdited>2024-10-03T11:55:51.6263082</lastEdited><Created>2023-12-06T11:08:16.5241063</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Hui</firstname><surname>Yao</surname><order>1</order></author><author><firstname>Yanhao</firstname><surname>Liu</surname><order>2</order></author><author><firstname>Haotian</firstname><surname>Lv</surname><order>3</order></author><author><firstname>Ju</firstname><surname>Huyan</surname><order>4</order></author><author><firstname>Zhanping</firstname><surname>You</surname><order>5</order></author><author><firstname>Yue</firstname><surname>Hou</surname><orcid>0000-0002-4334-2620</orcid><order>6</order></author></authors><documents><document><filename>65225__29221__4fad1e39c0a0481880f3f1ced99d90f3.pdf</filename><originalFilename>65225 VOR Yue HOU.pdf</originalFilename><uploaded>2023-12-06T11:42:25.4438875</uploaded><type>Output</type><contentLength>6285617</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>© 2023 The Authors. Distributed under the terms of a Creative Commons Attribution 4.0 International License (CC BY 4.0).</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling v2 65225 2023-12-06 Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition 92bf566c65343cb3ee04ad963eacf31b 0000-0002-4334-2620 Yue Hou Yue Hou true false 2023-12-06 ACEM Timely and accurate extraction of pavement crack information is crucial to maintain service conditions and structural safety for infrastructures and reduce further road maintenance costs. Currently, deep learning techniques for automated pavement crack detection are far superior to traditional manual approaches in both speed and accuracy. However, existing deep learning models may easily lose crack details when processing images containing complex background textures or other noises. Although many studies have alleviated this challenge by introducing attention mechanisms, especially the non-local (NL) block, which has the ability to efficiently capture long-range dependencies to facilitate crack pixel capture, the huge computational cost of NL makes the inference time of the model too long, which is not conducive to practical implementation. In this study, a new module, namely, the pyramid region attention module (PRAM), was developed by combining the pyramid pooling module in the pyramid scene parsing network and optimized NL, which can achieve global multi-scale context integration and long-range dependencies capture at a relatively lower computational cost. By applying PRAM to deep skip connections in the modified U-Net, an effective crack segmentation model called CrackResU-Net was developed. The test results on the existing CrackForest dataset showed that CrackResU-Net not only achieved an F1 score of 0.9580 but also took only 25.89 ms to process an image with a resolution of 480 × 320, which had advantages in accuracy and speed, compared with several other state-of-the-art crack segmentation approaches. It was fully demonstrated that this approach could realize automatic fast and high-precision recognition of pavement cracks for engineering purposes. Journal Article Computer-Aided Civil and Infrastructure Engineering 39 10 1490 1506 Wiley 1093-9687 1467-8667 15 5 2024 2024-05-15 10.1111/mice.13128 COLLEGE NANME Aerospace, Civil, Electrical, and Mechanical Engineering COLLEGE CODE ACEM Swansea University SU Library paid the OA fee (TA Institutional Deal) The authors appreciate the financial support from Hunan Expressway Group Co. Ltd and the Hunan Department of Transportation (No. 202152) in China. The authors also appreciate the funding support from the National Natural Science Foundation of China (No. 51778038), the China National Key R&D Program (2021YFB2600600, 2021YFB2600604, 2018YFB1600202), National Natural Science Foundation of China (grant number: 52108403). No. 202152, No. 51778038, 2021YFB2600600, 2021YFB2600604, 2018YFB1600202, grant number: 52108403 2024-10-03T11:55:51.6263082 2023-12-06T11:08:16.5241063 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Hui Yao 1 Yanhao Liu 2 Haotian Lv 3 Ju Huyan 4 Zhanping You 5 Yue Hou 0000-0002-4334-2620 6 65225__29221__4fad1e39c0a0481880f3f1ced99d90f3.pdf 65225 VOR Yue HOU.pdf 2023-12-06T11:42:25.4438875 Output 6285617 application/pdf Version of Record true © 2023 The Authors. Distributed under the terms of a Creative Commons Attribution 4.0 International License (CC BY 4.0). true eng https://creativecommons.org/licenses/by/4.0/
title Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition
spellingShingle Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition
Yue Hou
title_short Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition
title_full Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition
title_fullStr Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition
title_full_unstemmed Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition
title_sort Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition
author_id_str_mv 92bf566c65343cb3ee04ad963eacf31b
author_id_fullname_str_mv 92bf566c65343cb3ee04ad963eacf31b_***_Yue Hou
author Yue Hou
author2 Hui Yao
Yanhao Liu
Haotian Lv
Ju Huyan
Zhanping You
Yue Hou
format Journal article
container_title Computer-Aided Civil and Infrastructure Engineering
container_volume 39
container_issue 10
container_start_page 1490
publishDate 2024
institution Swansea University
issn 1093-9687
1467-8667
doi_str_mv 10.1111/mice.13128
publisher Wiley
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Timely and accurate extraction of pavement crack information is crucial to maintain service conditions and structural safety for infrastructures and reduce further road maintenance costs. Currently, deep learning techniques for automated pavement crack detection are far superior to traditional manual approaches in both speed and accuracy. However, existing deep learning models may easily lose crack details when processing images containing complex background textures or other noises. Although many studies have alleviated this challenge by introducing attention mechanisms, especially the non-local (NL) block, which has the ability to efficiently capture long-range dependencies to facilitate crack pixel capture, the huge computational cost of NL makes the inference time of the model too long, which is not conducive to practical implementation. In this study, a new module, namely, the pyramid region attention module (PRAM), was developed by combining the pyramid pooling module in the pyramid scene parsing network and optimized NL, which can achieve global multi-scale context integration and long-range dependencies capture at a relatively lower computational cost. By applying PRAM to deep skip connections in the modified U-Net, an effective crack segmentation model called CrackResU-Net was developed. The test results on the existing CrackForest dataset showed that CrackResU-Net not only achieved an F1 score of 0.9580 but also took only 25.89 ms to process an image with a resolution of 480 × 320, which had advantages in accuracy and speed, compared with several other state-of-the-art crack segmentation approaches. It was fully demonstrated that this approach could realize automatic fast and high-precision recognition of pavement cracks for engineering purposes.
published_date 2024-05-15T11:55:50Z
_version_ 1811889993281437696
score 11.036706

Similar Items