一种基于深度学习的方法,用于促进室内建筑工程的竣工状态识别

IF 3.1 Q2 CONSTRUCTION & BUILDING TECHNOLOGY Construction Innovation-England Pub Date : 2022-12-20 DOI:10.1108/ci-05-2022-0121
B. Ekanayake, Alireza Ahmadian Fard Fini, Johnny Kwok-Wai Wong, Peter Smith
{"title":"一种基于深度学习的方法,用于促进室内建筑工程的竣工状态识别","authors":"B. Ekanayake, Alireza Ahmadian Fard Fini, Johnny Kwok-Wai Wong, Peter Smith","doi":"10.1108/ci-05-2022-0121","DOIUrl":null,"url":null,"abstract":"\nPurpose\nRecognising the as-built state of construction elements is crucial for construction progress monitoring. Construction scholars have used computer vision-based algorithms to automate this process. Robust object recognition from indoor site images has been inhibited by technical challenges related to indoor objects, lighting conditions and camera positioning. Compared with traditional machine learning algorithms, one-stage detector deep learning (DL) algorithms can prioritise the inference speed, enable real-time accurate object detection and classification. This study aims to present a DL-based approach to facilitate the as-built state recognition of indoor construction works.\n\n\nDesign/methodology/approach\nThe one-stage DL-based approach was built upon YOLO version 4 (YOLOv4) algorithm using transfer learning with few hyperparameters customised and trained in the Google Colab virtual machine. The process of framing, insulation and drywall installation of indoor partitions was selected as the as-built scenario. For training, images were captured from two indoor sites with publicly available online images.\n\n\nFindings\nThe DL model reported a best-trained weight with a mean average precision of 92% and an average loss of 0.83. Compared to previous studies, the automation level of this study is high due to the use of fixed time-lapse cameras for data collection and zero manual intervention from the pre-processing algorithms to enhance visual quality of indoor images.\n\n\nOriginality/value\nThis study extends the application of DL models for recognising as-built state of indoor construction works upon providing training images. Presenting a workflow on training DL models in a virtual machine platform by reducing the computational complexities associated with DL models is also materialised.\n","PeriodicalId":45580,"journal":{"name":"Construction Innovation-England","volume":null,"pages":null},"PeriodicalIF":3.1000,"publicationDate":"2022-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"A deep learning-based approach to facilitate the as-built state recognition of indoor construction works\",\"authors\":\"B. Ekanayake, Alireza Ahmadian Fard Fini, Johnny Kwok-Wai Wong, Peter Smith\",\"doi\":\"10.1108/ci-05-2022-0121\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"\\nPurpose\\nRecognising the as-built state of construction elements is crucial for construction progress monitoring. Construction scholars have used computer vision-based algorithms to automate this process. Robust object recognition from indoor site images has been inhibited by technical challenges related to indoor objects, lighting conditions and camera positioning. Compared with traditional machine learning algorithms, one-stage detector deep learning (DL) algorithms can prioritise the inference speed, enable real-time accurate object detection and classification. This study aims to present a DL-based approach to facilitate the as-built state recognition of indoor construction works.\\n\\n\\nDesign/methodology/approach\\nThe one-stage DL-based approach was built upon YOLO version 4 (YOLOv4) algorithm using transfer learning with few hyperparameters customised and trained in the Google Colab virtual machine. The process of framing, insulation and drywall installation of indoor partitions was selected as the as-built scenario. For training, images were captured from two indoor sites with publicly available online images.\\n\\n\\nFindings\\nThe DL model reported a best-trained weight with a mean average precision of 92% and an average loss of 0.83. Compared to previous studies, the automation level of this study is high due to the use of fixed time-lapse cameras for data collection and zero manual intervention from the pre-processing algorithms to enhance visual quality of indoor images.\\n\\n\\nOriginality/value\\nThis study extends the application of DL models for recognising as-built state of indoor construction works upon providing training images. Presenting a workflow on training DL models in a virtual machine platform by reducing the computational complexities associated with DL models is also materialised.\\n\",\"PeriodicalId\":45580,\"journal\":{\"name\":\"Construction Innovation-England\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.1000,\"publicationDate\":\"2022-12-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Construction Innovation-England\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1108/ci-05-2022-0121\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"CONSTRUCTION & BUILDING TECHNOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Construction Innovation-England","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1108/ci-05-2022-0121","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CONSTRUCTION & BUILDING TECHNOLOGY","Score":null,"Total":0}
引用次数: 1

摘要

目的识别施工构件的竣工状态对于施工进度监控至关重要。建筑学者已经使用基于计算机视觉的算法来自动化这一过程。由于室内物体、照明条件和摄像机定位方面的技术挑战,室内场地图像的鲁棒物体识别受到了阻碍。与传统的机器学习算法相比,一阶段检测器深度学习(DL)算法可以优先考虑推理速度,实现实时准确的目标检测和分类。本研究旨在提出一种基于DL的方法,以促进室内建筑工程的竣工状态识别。设计/方法论/方法基于DL的一阶段方法建立在YOLO版本4(YOLOv4)算法的基础上,使用迁移学习,在Google Colab虚拟机中定制和训练很少的超参数。室内隔墙的框架、隔热和干墙安装过程被选为竣工场景。为了训练,从两个室内网站拍摄了图像,并公开了在线图像。发现DL模型报告了最佳训练重量,平均精度为92%,平均损失为0.83。与以往的研究相比,本研究的自动化水平较高,因为使用了固定的延时相机进行数据收集,预处理算法零人工干预,以提高室内图像的视觉质量。独创性/价值本研究扩展了DL模型在提供训练图像后识别室内建筑工程竣工状态的应用。还实现了通过降低与DL模型相关联的计算复杂性来在虚拟机平台中呈现关于训练DL模型的工作流程。
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
A deep learning-based approach to facilitate the as-built state recognition of indoor construction works
Purpose Recognising the as-built state of construction elements is crucial for construction progress monitoring. Construction scholars have used computer vision-based algorithms to automate this process. Robust object recognition from indoor site images has been inhibited by technical challenges related to indoor objects, lighting conditions and camera positioning. Compared with traditional machine learning algorithms, one-stage detector deep learning (DL) algorithms can prioritise the inference speed, enable real-time accurate object detection and classification. This study aims to present a DL-based approach to facilitate the as-built state recognition of indoor construction works. Design/methodology/approach The one-stage DL-based approach was built upon YOLO version 4 (YOLOv4) algorithm using transfer learning with few hyperparameters customised and trained in the Google Colab virtual machine. The process of framing, insulation and drywall installation of indoor partitions was selected as the as-built scenario. For training, images were captured from two indoor sites with publicly available online images. Findings The DL model reported a best-trained weight with a mean average precision of 92% and an average loss of 0.83. Compared to previous studies, the automation level of this study is high due to the use of fixed time-lapse cameras for data collection and zero manual intervention from the pre-processing algorithms to enhance visual quality of indoor images. Originality/value This study extends the application of DL models for recognising as-built state of indoor construction works upon providing training images. Presenting a workflow on training DL models in a virtual machine platform by reducing the computational complexities associated with DL models is also materialised.
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
Construction Innovation-England
Construction Innovation-England CONSTRUCTION & BUILDING TECHNOLOGY-
CiteScore
7.10
自引率
12.10%
发文量
71
期刊最新文献
Usage of digital technology in improving the mental health of workers on construction sites Uncertainties affecting the offsite construction supply chain resilience: a systematic literature review Developing an interactive pile training module for construction risk management and gaging users’ intentions Impact of trust in virtual project teams: structural equation modelling approach Customized shading solutions for complex building façades: the potential of an innovative cement-textile composite material through a performance-based generative design
×
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