A state-of-the-art survey of deep learning models for automated pavement crack segmentation

IF 4.3 Q2 TRANSPORTATION
Hongren Gong, Liming Liu, Haimei Liang, Yuhui Zhou, Lin Cong
{"title":"A state-of-the-art survey of deep learning models for automated pavement crack segmentation","authors":"Hongren Gong,&nbsp;Liming Liu,&nbsp;Haimei Liang,&nbsp;Yuhui Zhou,&nbsp;Lin Cong","doi":"10.1016/j.ijtst.2023.11.005","DOIUrl":null,"url":null,"abstract":"<div><p>Survey of road cracks in a timely, complete, and accurate way is pivotal to pavement maintenance planning. Motivated by the increasingly heavy task of identifying cracks, researchers have developed extensive crack segmentation models based on Deep learning (DL) methods with significantly different levels of accuracy, efficiency, and generalizing capacity. Although many of the models provide satisfying detection performance, why these models work still needs to be determined. The objective of this study is to survey recent advances in automated DL crack recognition and provide evidence for their underlying working mechanism. We first reviewed 54 DL crack recognition methods to summarize critical factors in these models. Then, we conducted a performance evaluation of fourteen famous semantic segmentation models using the quantitative metrics: F-1 score and mIoU. Then, the effective receptive field and class activation map of the included models are visualized to demonstrate the training results as qualitative evaluation. Based on the literature review and comparison results, larger kernel size, feature fusion, and attention module all contribute to the improvement of model performance. Striking a balance between increasing the effective receptive field and computational/memory efficiency is the key to designing DL crack segmentation models. Finally, some potential directions and suggestions for future development are provided, such as developing semi-supervised or unsupervised learning for the high cost of pixel-level labeling.</p></div>","PeriodicalId":52282,"journal":{"name":"International Journal of Transportation Science and Technology","volume":"13 ","pages":"Pages 44-57"},"PeriodicalIF":4.3000,"publicationDate":"2023-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2046043023001028/pdfft?md5=6e19d13a3fcc3f859e9440b5816c4981&pid=1-s2.0-S2046043023001028-main.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Transportation Science and Technology","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2046043023001028","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"TRANSPORTATION","Score":null,"Total":0}
引用次数: 0

Abstract

Survey of road cracks in a timely, complete, and accurate way is pivotal to pavement maintenance planning. Motivated by the increasingly heavy task of identifying cracks, researchers have developed extensive crack segmentation models based on Deep learning (DL) methods with significantly different levels of accuracy, efficiency, and generalizing capacity. Although many of the models provide satisfying detection performance, why these models work still needs to be determined. The objective of this study is to survey recent advances in automated DL crack recognition and provide evidence for their underlying working mechanism. We first reviewed 54 DL crack recognition methods to summarize critical factors in these models. Then, we conducted a performance evaluation of fourteen famous semantic segmentation models using the quantitative metrics: F-1 score and mIoU. Then, the effective receptive field and class activation map of the included models are visualized to demonstrate the training results as qualitative evaluation. Based on the literature review and comparison results, larger kernel size, feature fusion, and attention module all contribute to the improvement of model performance. Striking a balance between increasing the effective receptive field and computational/memory efficiency is the key to designing DL crack segmentation models. Finally, some potential directions and suggestions for future development are provided, such as developing semi-supervised or unsupervised learning for the high cost of pixel-level labeling.

路面裂缝自动分割深度学习模型的最新研究成果
及时、完整、准确地调查路面裂缝对路面维护规划至关重要。在日益繁重的裂缝识别任务的推动下,研究人员基于深度学习(DL)方法开发了大量的裂缝分割模型,这些模型在准确性、效率和泛化能力方面都有很大的不同。尽管许多模型都能提供令人满意的检测性能,但这些模型为何能发挥作用仍有待确定。本研究的目的是调查最近在自动 DL 裂纹识别方面取得的进展,并为其基本工作机制提供证据。我们首先回顾了 54 种 DL 裂纹识别方法,总结了这些模型中的关键因素。然后,我们使用量化指标对十四种著名的语义分割模型进行了性能评估:F-1 分数和 mIoU。然后,我们将所包含模型的有效感受野和类激活图可视化,以展示训练结果作为定性评估。根据文献综述和比较结果,更大的核大小、特征融合和注意力模块都有助于提高模型性能。在提高有效感受野和计算/内存效率之间取得平衡是设计 DL 裂缝分割模型的关键。最后,我们还提供了一些未来发展的潜在方向和建议,例如针对像素级标记的高成本开发半监督或无监督学习。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
International Journal of Transportation Science and Technology
International Journal of Transportation Science and Technology Engineering-Civil and Structural Engineering
CiteScore
7.20
自引率
0.00%
发文量
105
审稿时长
88 days
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信