基于深度学习的裂缝分割和裂缝扩展快速评估研究

IF 2.9 3区 工程技术 Q2 ENGINEERING, CIVIL Frontiers of Structural and Civil Engineering Pub Date : 2024-05-30 DOI:10.1007/s11709-024-1040-z
Than V. Tran, H. Nguyen-Xuan, Xiaoying Zhuang
{"title":"基于深度学习的裂缝分割和裂缝扩展快速评估研究","authors":"Than V. Tran, H. Nguyen-Xuan, Xiaoying Zhuang","doi":"10.1007/s11709-024-1040-z","DOIUrl":null,"url":null,"abstract":"<p>Identifying crack and predicting crack propagation are critical processes for the risk assessment of engineering structures. Most traditional approaches to crack modeling are faced with issues of high computational costs and excessive computing time. To address this issue, we explore the potential of deep learning (DL) to increase the efficiency of crack detection and forecasting crack growth. However, there is no single algorithm that can fit all data sets well or can apply in all cases since specific tasks vary. In the paper, we present DL models for identifying cracks, especially on concrete surface images, and for predicting crack propagation. Firstly, SegNet and U-Net networks are used to identify concrete cracks. Stochastic gradient descent (SGD) and adaptive moment estimation (Adam) algorithms are applied to minimize loss function during iterations. Secondly, time series algorithms including gated recurrent unit (GRU) and long short-term memory (LSTM) are used to predict crack propagation. The experimental findings indicate that the U-Net is more robust and efficient than the SegNet for identifying crack segmentation and achieves the most outstanding results. For evaluation of crack propagation, GRU and LSTM are used as DL models and results show good agreement with the experimental data.</p>","PeriodicalId":12476,"journal":{"name":"Frontiers of Structural and Civil Engineering","volume":"44 1","pages":""},"PeriodicalIF":2.9000,"publicationDate":"2024-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Investigation of crack segmentation and fast evaluation of crack propagation, based on deep learning\",\"authors\":\"Than V. Tran, H. Nguyen-Xuan, Xiaoying Zhuang\",\"doi\":\"10.1007/s11709-024-1040-z\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Identifying crack and predicting crack propagation are critical processes for the risk assessment of engineering structures. Most traditional approaches to crack modeling are faced with issues of high computational costs and excessive computing time. To address this issue, we explore the potential of deep learning (DL) to increase the efficiency of crack detection and forecasting crack growth. However, there is no single algorithm that can fit all data sets well or can apply in all cases since specific tasks vary. In the paper, we present DL models for identifying cracks, especially on concrete surface images, and for predicting crack propagation. Firstly, SegNet and U-Net networks are used to identify concrete cracks. Stochastic gradient descent (SGD) and adaptive moment estimation (Adam) algorithms are applied to minimize loss function during iterations. Secondly, time series algorithms including gated recurrent unit (GRU) and long short-term memory (LSTM) are used to predict crack propagation. The experimental findings indicate that the U-Net is more robust and efficient than the SegNet for identifying crack segmentation and achieves the most outstanding results. For evaluation of crack propagation, GRU and LSTM are used as DL models and results show good agreement with the experimental data.</p>\",\"PeriodicalId\":12476,\"journal\":{\"name\":\"Frontiers of Structural and Civil Engineering\",\"volume\":\"44 1\",\"pages\":\"\"},\"PeriodicalIF\":2.9000,\"publicationDate\":\"2024-05-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Frontiers of Structural and Civil Engineering\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1007/s11709-024-1040-z\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, CIVIL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers of Structural and Civil Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1007/s11709-024-1040-z","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, CIVIL","Score":null,"Total":0}
引用次数: 0

摘要

识别裂纹和预测裂纹扩展是工程结构风险评估的关键过程。大多数传统的裂纹建模方法都面临着计算成本高和计算时间过长的问题。为了解决这个问题,我们探索了深度学习(DL)的潜力,以提高裂纹检测和预测裂纹生长的效率。然而,由于具体任务各不相同,没有一种算法能很好地适应所有数据集或适用于所有情况。本文介绍了用于识别裂缝(尤其是混凝土表面图像)和预测裂缝扩展的 DL 模型。首先,我们使用 SegNet 和 U-Net 网络来识别混凝土裂缝。在迭代过程中,采用随机梯度下降(SGD)和自适应矩估计(Adam)算法来最小化损失函数。其次,使用时间序列算法,包括门控递归单元(GRU)和长短期记忆(LSTM)来预测裂缝的扩展。实验结果表明,在识别裂缝分割方面,U-Net 比 SegNet 更稳健、更高效,并取得了最出色的结果。在评估裂纹传播时,使用 GRU 和 LSTM 作为 DL 模型,结果显示与实验数据有很好的一致性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Investigation of crack segmentation and fast evaluation of crack propagation, based on deep learning

Identifying crack and predicting crack propagation are critical processes for the risk assessment of engineering structures. Most traditional approaches to crack modeling are faced with issues of high computational costs and excessive computing time. To address this issue, we explore the potential of deep learning (DL) to increase the efficiency of crack detection and forecasting crack growth. However, there is no single algorithm that can fit all data sets well or can apply in all cases since specific tasks vary. In the paper, we present DL models for identifying cracks, especially on concrete surface images, and for predicting crack propagation. Firstly, SegNet and U-Net networks are used to identify concrete cracks. Stochastic gradient descent (SGD) and adaptive moment estimation (Adam) algorithms are applied to minimize loss function during iterations. Secondly, time series algorithms including gated recurrent unit (GRU) and long short-term memory (LSTM) are used to predict crack propagation. The experimental findings indicate that the U-Net is more robust and efficient than the SegNet for identifying crack segmentation and achieves the most outstanding results. For evaluation of crack propagation, GRU and LSTM are used as DL models and results show good agreement with the experimental data.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
5.20
自引率
3.30%
发文量
734
期刊介绍: Frontiers of Structural and Civil Engineering is an international journal that publishes original research papers, review articles and case studies related to civil and structural engineering. Topics include but are not limited to the latest developments in building and bridge structures, geotechnical engineering, hydraulic engineering, coastal engineering, and transport engineering. Case studies that demonstrate the successful applications of cutting-edge research technologies are welcome. The journal also promotes and publishes interdisciplinary research and applications connecting civil engineering and other disciplines, such as bio-, info-, nano- and social sciences and technology. Manuscripts submitted for publication will be subject to a stringent peer review.
期刊最新文献
An artificial neural network based deep collocation method for the solution of transient linear and nonlinear partial differential equations Bibliographic survey and comprehensive review on mechanical and durability properties of microorganism based self-healing concrete Seismic response of pile-supported structures considering the coupling of inertial and kinematic interactions in different soil sites An isogeometric approach for nonlocal bending and free oscillation of magneto-electro-elastic functionally graded nanobeam with elastic constraints Shaking table test on a tunnel-group metro station in rock site under harmonic excitation
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1