首页 > 最新文献

Automation in Construction最新文献

英文 中文
Real-time lightweight YOLO model for grouting defect detection in external post-tensioned ducts via infrared thermography 通过红外热成像检测外部后张法管道灌浆缺陷的实时轻量级 YOLO 模型
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-21 DOI: 10.1016/j.autcon.2024.105830
It is challenging to distinguish the defective areas using infrared thermography to automatically analyze external post-tensioned tendon duct grouting defects. To achieve efficient and stable automated detection, a lightweight real-time grouting defects detection method based on YOLO deep learning is proposed. Firstly, the Cutpaste data augmentation method was used to effectively alleviate the problem of overfitting. Then, the C3Ghost module was introduced into the neck network, and the number of channels in the network layers was adjusted to 50 % of those in the YOLOv5s model, reducing the number of parameters and computational resources. Finally, the SGD optimizer and GIOU loss function, as well as the Sim attention module, were used to improve detection accuracy. Based on instance analysis and comparison, this method achieves [email protected] of 96.9 % and detection speed of 66FPS. Compared with YOLOv5s, it reduces the number of parameters by 79 % and FLOPs by 77 %.
利用红外热成像技术自动分析外部后张法筋管灌浆缺陷,区分缺陷区域是一项挑战。为了实现高效稳定的自动检测,本文提出了一种基于 YOLO 深度学习的轻量级灌浆缺陷实时检测方法。首先,使用 Cutpaste 数据增强方法有效缓解了过拟合问题。然后,在颈部网络中引入 C3Ghost 模块,并将网络层的通道数调整为 YOLOv5s 模型的 50%,减少了参数数量和计算资源。最后,利用 SGD 优化器和 GIOU 损失函数以及 Sim attention 模块提高了检测精度。根据实例分析和比较,该方法的[email protected] 达到 96.9 %,检测速度达到 66FPS。与 YOLOv5s 相比,参数数量减少了 79%,FLOPs 减少了 77%。
{"title":"Real-time lightweight YOLO model for grouting defect detection in external post-tensioned ducts via infrared thermography","authors":"","doi":"10.1016/j.autcon.2024.105830","DOIUrl":"10.1016/j.autcon.2024.105830","url":null,"abstract":"<div><div>It is challenging to distinguish the defective areas using infrared thermography to automatically analyze external post-tensioned tendon duct grouting defects. To achieve efficient and stable automated detection, a lightweight real-time grouting defects detection method based on YOLO deep learning is proposed. Firstly, the Cutpaste data augmentation method was used to effectively alleviate the problem of overfitting. Then, the C3Ghost module was introduced into the neck network, and the number of channels in the network layers was adjusted to 50 % of those in the YOLOv5s model, reducing the number of parameters and computational resources. Finally, the SGD optimizer and GIOU loss function, as well as the Sim attention module, were used to improve detection accuracy. Based on instance analysis and comparison, this method achieves [email protected] of 96.9 % and detection speed of 66FPS. Compared with YOLOv5s, it reduces the number of parameters by 79 % and FLOPs by 77 %.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142534918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Paving block displacement detection and measurement using 3D laser sensors on unmanned ground vehicles 利用无人地面车辆上的 3D 激光传感器检测和测量铺路块的位移
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-17 DOI: 10.1016/j.autcon.2024.105813
Construction sites with deep excavation in urban areas can induce ground deformation, potentially harming nearby infrastructure. Therefore, monitoring construction sites is crucial. Typically, a sidewalk is located adjacent to the construction site, and ground deformation can cause the displacement of paving blocks. Accurate measurement of paving block displacement and cracks is essential. This paper proposes an efficient automated detection and measurement method using a 3D laser line sensor on Unmanned Ground Vehicles (UGVs), emphasizing online measurement capabilities. The method involves two steps: detecting target objects via 2D projection from 3D point cloud data and measuring object features by reducing unnecessary data with the Clustered Piecewise Linear Fitting (CPLF) algorithm. This two-step process enhances parallelism between edge servers and devices, thereby reducing total processing time. Prototype implementation and experiments show that our method achieves low errors of accuracy and is suitable for automated online detection and measurement on UGVs.
在城市地区进行深度挖掘的建筑工地可能会引起地面变形,对附近的基础设施造成潜在危害。因此,对施工现场进行监测至关重要。通常情况下,人行道位于施工现场附近,地面变形会导致铺路砖位移。准确测量铺路块的位移和裂缝至关重要。本文利用无人地面车辆(UGV)上的三维激光线传感器,提出了一种高效的自动探测和测量方法,强调在线测量功能。该方法包括两个步骤:通过三维点云数据的二维投影检测目标物体,以及通过聚类分片线性拟合(CPLF)算法减少不必要的数据来测量物体特征。这两个步骤增强了边缘服务器和设备之间的并行性,从而缩短了总处理时间。原型实施和实验表明,我们的方法实现了较低的精度误差,适用于 UGV 的自动在线检测和测量。
{"title":"Paving block displacement detection and measurement using 3D laser sensors on unmanned ground vehicles","authors":"","doi":"10.1016/j.autcon.2024.105813","DOIUrl":"10.1016/j.autcon.2024.105813","url":null,"abstract":"<div><div>Construction sites with deep excavation in urban areas can induce ground deformation, potentially harming nearby infrastructure. Therefore, monitoring construction sites is crucial. Typically, a sidewalk is located adjacent to the construction site, and ground deformation can cause the displacement of paving blocks. Accurate measurement of paving block displacement and cracks is essential. This paper proposes an efficient automated detection and measurement method using a 3D laser line sensor on Unmanned Ground Vehicles (UGVs), emphasizing online measurement capabilities. The method involves two steps: detecting target objects via 2D projection from 3D point cloud data and measuring object features by reducing unnecessary data with the Clustered Piecewise Linear Fitting (CPLF) algorithm. This two-step process enhances parallelism between edge servers and devices, thereby reducing total processing time. Prototype implementation and experiments show that our method achieves low errors of accuracy and is suitable for automated online detection and measurement on UGVs.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142446980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Impact of environmental pollutants on work performance using virtual reality 利用虚拟现实技术研究环境污染物对工作表现的影响
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-17 DOI: 10.1016/j.autcon.2024.105833
Virtual reality-based experiments were conducted to assess the impacts of environmental pollutants (i.e., noise, vibration, and dust) on work performance. In these experiments, concrete chipping work was performed in eight different exposure environments based on exposure to three environmental pollutants to measure data related to work performance: (i) work performance metrics, including work duration and accuracy; and (ii) mental workload. The relationships between data related to work performance and environmental pollutants were then analyzed using statistical techniques as follows: First, work duration was statistically significantly affected by dust, while work accuracy was significantly affected by vibration. Second, mental workload was statistically significantly affected by all three environmental pollutants, increasing with the number of environmental pollutants the workers exposed to. Third, all data related to work performance were found to be correlated with each other. These findings provide insights into improving work performance by managing environmental pollutants in the construction industry.
进行了基于虚拟现实的实验,以评估环境污染物(即噪音、振动和粉尘)对工作绩效的影响。在这些实验中,根据三种环境污染物的暴露情况,在八种不同的暴露环境中进行了混凝土削削工作,以测量与工作绩效相关的数据:(i) 工作绩效指标,包括工作持续时间和准确性;以及 (ii) 精神工作量。然后使用统计技术分析了工作绩效相关数据与环境污染物之间的关系,具体如下:首先,从统计学角度看,粉尘对工作持续时间有显著影响,而振动对工作准确性有显著影响。其次,从统计学角度看,脑力工作量受所有三种环境污染物的影响都很大,并且随着工人接触的环境污染物数量的增加而增加。第三,所有与工作绩效相关的数据都相互关联。这些发现为通过管理建筑行业的环境污染物来提高工作绩效提供了启示。
{"title":"Impact of environmental pollutants on work performance using virtual reality","authors":"","doi":"10.1016/j.autcon.2024.105833","DOIUrl":"10.1016/j.autcon.2024.105833","url":null,"abstract":"<div><div>Virtual reality-based experiments were conducted to assess the impacts of environmental pollutants (i.e., noise, vibration, and dust) on work performance. In these experiments, concrete chipping work was performed in eight different exposure environments based on exposure to three environmental pollutants to measure data related to work performance: (i) work performance metrics, including work duration and accuracy; and (ii) mental workload. The relationships between data related to work performance and environmental pollutants were then analyzed using statistical techniques as follows: First, work duration was statistically significantly affected by dust, while work accuracy was significantly affected by vibration. Second, mental workload was statistically significantly affected by all three environmental pollutants, increasing with the number of environmental pollutants the workers exposed to. Third, all data related to work performance were found to be correlated with each other. These findings provide insights into improving work performance by managing environmental pollutants in the construction industry.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142446979","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Artificial intelligence driven tunneling-induced surface settlement prediction 人工智能驱动的隧道诱导地表沉降预测
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-17 DOI: 10.1016/j.autcon.2024.105819
There has been an increasing demand for shield tunnel construction due to the extensive utilization and limited land in metropolitan cities. However, the behaviors of soils and rocks exhibit a high level of uncertainty in material modeling. Artificial Intelligence (AI) techniques exhibit huge potential in addressing geotechnical issues that involve non-linear soil-structure interaction. This paper aims to review AI-driven prediction of tunneling-induced surface settlement, focusing on aspects of dataset establishment, input feature selection, and hyperparameter optimization. An overview of AI key applications in surface settlement prediction over the past decades is compiled. Furthermore, the capabilities and limitations of diverse AI techniques are discussed, guiding the selection of methodologies for different scenarios. Subsequently, recent developments such as AI variants, the latest optimization algorithms, and cutting-edge methods are illustrated. Lastly, possible countermeasures of AI for challenges in pragmatic applications are proposed, offering orientations for further research in AI-driven tunneling-induced surface settlement prediction.
由于大都市的广泛利用和土地有限,对盾构隧道建设的需求与日俱增。然而,在材料建模中,土壤和岩石的行为具有很大的不确定性。人工智能(AI)技术在解决涉及非线性土-结构相互作用的岩土工程问题方面展现出巨大的潜力。本文旨在回顾人工智能驱动的隧道诱发地表沉降预测,重点关注数据集建立、输入特征选择和超参数优化等方面。本文概述了过去几十年来人工智能在地表沉降预测中的主要应用。此外,还讨论了各种人工智能技术的能力和局限性,为不同情况下的方法选择提供指导。随后,阐述了人工智能变体、最新优化算法和前沿方法等最新发展。最后,针对人工智能在实际应用中面临的挑战提出了可能的对策,为人工智能驱动的隧道诱导地表沉降预测的进一步研究提供了方向。
{"title":"Artificial intelligence driven tunneling-induced surface settlement prediction","authors":"","doi":"10.1016/j.autcon.2024.105819","DOIUrl":"10.1016/j.autcon.2024.105819","url":null,"abstract":"<div><div>There has been an increasing demand for shield tunnel construction due to the extensive utilization and limited land in metropolitan cities. However, the behaviors of soils and rocks exhibit a high level of uncertainty in material modeling. Artificial Intelligence (AI) techniques exhibit huge potential in addressing geotechnical issues that involve non-linear soil-structure interaction. This paper aims to review AI-driven prediction of tunneling-induced surface settlement, focusing on aspects of dataset establishment, input feature selection, and hyperparameter optimization. An overview of AI key applications in surface settlement prediction over the past decades is compiled. Furthermore, the capabilities and limitations of diverse AI techniques are discussed, guiding the selection of methodologies for different scenarios. Subsequently, recent developments such as AI variants, the latest optimization algorithms, and cutting-edge methods are illustrated. Lastly, possible countermeasures of AI for challenges in pragmatic applications are proposed, offering orientations for further research in AI-driven tunneling-induced surface settlement prediction.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142446981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Automated progress monitoring of land development projects using unmanned aerial vehicles and machine learning 利用无人飞行器和机器学习自动监测土地开发项目的进度
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-16 DOI: 10.1016/j.autcon.2024.105827
In land development projects, effective control of the engineering progress is crucial for managing construction quality and costs. However, the conventional approach to monitoring progress is inadequate for large-scale projects. This paper proposes a technique that utilizes UAV images and machine learning techniques to monitor land development projects. The object detection and image segmentation techniques were used to detect essential construction objects. The detected objects were automatically compared to design drawings for checking the progress of the project. Moreover, to ensure personnel safety during construction, an automated process for identifying locations requiring safety barriers was also designed in the study. The effectiveness of all the proposed techniques was evaluated in a real case study. It is illustrated that this fully automated approach for land development monitoring is efficient and thus can contribute to construction safety, cost reduction, and quality assurance in a land development project.
在土地开发项目中,有效控制工程进度对于管理施工质量和成本至关重要。然而,传统的进度监控方法并不适用于大型项目。本文提出了一种利用无人机图像和机器学习技术监控土地开发项目的技术。利用物体检测和图像分割技术来检测重要的施工物体。检测到的物体会自动与设计图纸进行比对,以检查项目进度。此外,为确保施工期间的人员安全,研究还设计了一个自动流程,用于识别需要安全屏障的位置。在实际案例研究中对所有建议技术的有效性进行了评估。结果表明,这种用于土地开发监测的全自动方法非常高效,因此有助于土地开发项目中的施工安全、成本降低和质量保证。
{"title":"Automated progress monitoring of land development projects using unmanned aerial vehicles and machine learning","authors":"","doi":"10.1016/j.autcon.2024.105827","DOIUrl":"10.1016/j.autcon.2024.105827","url":null,"abstract":"<div><div>In land development projects, effective control of the engineering progress is crucial for managing construction quality and costs. However, the conventional approach to monitoring progress is inadequate for large-scale projects. This paper proposes a technique that utilizes UAV images and machine learning techniques to monitor land development projects. The object detection and image segmentation techniques were used to detect essential construction objects. The detected objects were automatically compared to design drawings for checking the progress of the project. Moreover, to ensure personnel safety during construction, an automated process for identifying locations requiring safety barriers was also designed in the study. The effectiveness of all the proposed techniques was evaluated in a real case study. It is illustrated that this fully automated approach for land development monitoring is efficient and thus can contribute to construction safety, cost reduction, and quality assurance in a land development project.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142442010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Weakly supervised 3D point cloud semantic segmentation for architectural heritage using teacher-guided consistency and contrast learning 利用教师指导的一致性和对比度学习对建筑遗产进行弱监督三维点云语义分割
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-16 DOI: 10.1016/j.autcon.2024.105831
Point cloud semantic segmentation is significant for managing and protecting architectural heritage. Currently, fully supervised methods require a large amount of annotated data, while weakly supervised methods are difficult to transfer directly to architectural heritage. This paper proposes an end-to-end teacher-guided consistency and contrastive learning weakly supervised (TCCWS) framework for architectural heritage point cloud semantic segmentation, which can fully utilize limited labeled points to train network. Specifically, a teacher-student framework is designed to generate pseudo labels and a pseudo label dividing module is proposed to distinguish reliable and ambiguous point sets. Based on it, a consistency and contrastive learning strategy is designed to fully utilize supervision signals to learn the features of point clouds. The framework is tested on the ArCH dataset and self-collected point cloud, which demonstrates that the proposed method can achieve effective semantic segmentation of architectural heritage using only 0.1 % of annotated points.
点云语义分割对管理和保护建筑遗产意义重大。目前,全监督方法需要大量注释数据,而弱监督方法很难直接应用于建筑遗产。本文针对建筑遗产点云语义分割提出了端到端的教师指导一致性和对比学习弱监督(TCCWS)框架,可充分利用有限的标注点来训练网络。具体来说,设计了一个师生框架来生成伪标签,并提出了一个伪标签划分模块来区分可靠点集和模糊点集。在此基础上,设计了一种一致性和对比性学习策略,充分利用监督信号来学习点云的特征。该框架在 ArCH 数据集和自采集点云上进行了测试,结果表明,所提出的方法只需使用 0.1% 的注释点就能实现对建筑遗产的有效语义分割。
{"title":"Weakly supervised 3D point cloud semantic segmentation for architectural heritage using teacher-guided consistency and contrast learning","authors":"","doi":"10.1016/j.autcon.2024.105831","DOIUrl":"10.1016/j.autcon.2024.105831","url":null,"abstract":"<div><div>Point cloud semantic segmentation is significant for managing and protecting architectural heritage. Currently, fully supervised methods require a large amount of annotated data, while weakly supervised methods are difficult to transfer directly to architectural heritage. This paper proposes an end-to-end teacher-guided consistency and contrastive learning weakly supervised (TCCWS) framework for architectural heritage point cloud semantic segmentation, which can fully utilize limited labeled points to train network. Specifically, a teacher-student framework is designed to generate pseudo labels and a pseudo label dividing module is proposed to distinguish reliable and ambiguous point sets. Based on it, a consistency and contrastive learning strategy is designed to fully utilize supervision signals to learn the features of point clouds. The framework is tested on the ArCH dataset and self-collected point cloud, which demonstrates that the proposed method can achieve effective semantic segmentation of architectural heritage using only 0.1 % of annotated points.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142442011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Fully automated extraction of railtop centerline from mobile laser scanning data 从移动激光扫描数据中全自动提取轨顶中心线
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-16 DOI: 10.1016/j.autcon.2024.105812
Digitization is an important part of efficient infrastructure maintenance. Means to achieve a digital asset database include precise 3D surveys of the physical assets and advanced automated recognition of objects of interest for documenting, maintenance and further analysis purposes. To this end, fast data collection of railway infrastructure environments can be obtained using a mobile laser scanner mounted on a service locomotive, permitting uninterruptive service. This paper presents an algorithm that extracts the railtop centerlines of up to seven parallel tracks with a single measurement pass and achieves an accuracy of 0.3 cm to 0.8 cm on non-intersecting rails, which improves the state of the art by 55%–85%. On intersecting rails, the railtop location accuracy is comparable to that of existing methods. The proposed method uses only geometric data and performs in real time in two-track railroad configurations.
数字化是高效基础设施维护的重要组成部分。实现数字化资产数据库的方法包括对实物资产进行精确的三维测量,以及先进的自动识别相关对象,以达到记录、维护和进一步分析的目的。为此,可使用安装在服务机车上的移动激光扫描仪对铁路基础设施环境进行快速数据采集,从而实现不间断服务。本文介绍了一种算法,只需一次测量就能提取多达七条平行轨道的轨顶中心线,在不相交的轨道上可达到 0.3 厘米至 0.8 厘米的精度,比现有技术提高了 55% 至 85%。在相交轨道上,轨顶定位精度与现有方法相当。建议的方法仅使用几何数据,可在双轨铁路配置中实时执行。
{"title":"Fully automated extraction of railtop centerline from mobile laser scanning data","authors":"","doi":"10.1016/j.autcon.2024.105812","DOIUrl":"10.1016/j.autcon.2024.105812","url":null,"abstract":"<div><div>Digitization is an important part of efficient infrastructure maintenance. Means to achieve a digital asset database include precise 3D surveys of the physical assets and advanced automated recognition of objects of interest for documenting, maintenance and further analysis purposes. To this end, fast data collection of railway infrastructure environments can be obtained using a mobile laser scanner mounted on a service locomotive, permitting uninterruptive service. This paper presents an algorithm that extracts the railtop centerlines of up to seven parallel tracks with a single measurement pass and achieves an accuracy of 0.3<!--> <!-->cm to 0.8<!--> <!-->cm on non-intersecting rails, which improves the state of the art by 55%–85%. On intersecting rails, the railtop location accuracy is comparable to that of existing methods. The proposed method uses only geometric data and performs in real time in two-track railroad configurations.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142441930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Generation of LOD4 models for buildings towards the automated 3D modeling of BIMs and digital twins 生成建筑物 LOD4 模型,实现 BIM 和数字双胞胎的自动 3D 建模
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-16 DOI: 10.1016/j.autcon.2024.105822
An image-based methodology is presented for the automatic generation of geometric building models at LOD4, incorporating both interior and exterior geometrical information. Existing approaches often focus on simplified geometries for either exteriors or interiors, leading to integration challenges due to data complexity and processing demands. This methodology addresses these challenges by utilizing three structure-from-motion models: one for the building exterior, one for the interior, and one for the entrance. The exterior and interior data are processed separately using planar primitives, and the models are subsequently aligned through a 3D point cloud registration method based on 2D image features. This ensures a unified coordinate system and accurate generation of the LOD4 model. The framework achieved a mean relative error of 3.06% and a mean absolute error of 0.05 m, underscoring its robustness for applications such as numerical modeling, construction management, and structural health monitoring, making it valuable for further advancements in building information models and digital twins.
本文介绍了一种基于图像的方法,用于自动生成 LOD4 级的几何建筑模型,其中包含内部和外部的几何信息。现有的方法通常侧重于外部或内部的简化几何图形,由于数据的复杂性和处理需求,导致集成方面的挑战。本方法通过使用三个运动结构模型来解决这些难题:一个是建筑外部模型,一个是建筑内部模型,还有一个是建筑入口模型。外部和内部数据分别使用平面基元进行处理,然后通过基于二维图像特征的三维点云注册方法对这些模型进行对齐。这确保了统一的坐标系和 LOD4 模型的精确生成。该框架的平均相对误差为 3.06%,平均绝对误差为 0.05 米,突出了其在数值建模、施工管理和结构健康监测等应用中的稳健性,使其在进一步推进建筑信息模型和数字孪生中具有重要价值。
{"title":"Generation of LOD4 models for buildings towards the automated 3D modeling of BIMs and digital twins","authors":"","doi":"10.1016/j.autcon.2024.105822","DOIUrl":"10.1016/j.autcon.2024.105822","url":null,"abstract":"<div><div>An image-based methodology is presented for the automatic generation of geometric building models at LOD4, incorporating both interior and exterior geometrical information. Existing approaches often focus on simplified geometries for either exteriors or interiors, leading to integration challenges due to data complexity and processing demands. This methodology addresses these challenges by utilizing three structure-from-motion models: one for the building exterior, one for the interior, and one for the entrance. The exterior and interior data are processed separately using planar primitives, and the models are subsequently aligned through a 3D point cloud registration method based on 2D image features. This ensures a unified coordinate system and accurate generation of the LOD4 model. The framework achieved a mean relative error of 3.06% and a mean absolute error of 0.05 m, underscoring its robustness for applications such as numerical modeling, construction management, and structural health monitoring, making it valuable for further advancements in building information models and digital twins.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142441929","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Quantitative assessment of cracks in concrete structures using active-learning-integrated transformer and unmanned robotic platform 利用主动学习集成变压器和无人机器人平台对混凝土结构裂缝进行定量评估
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-15 DOI: 10.1016/j.autcon.2024.105829
Quantitative assessment of cracks in concrete bridges is crucial for structural health monitoring and digital twin. However, the training of crack segmentation models relies heavily on annotation resources, and their segmentation capabilities are often unsatisfactory in terms of the accuracy of boundary location of thin cracks encountered in practice. In this paper, an active-learning-integrated crack segmentation transformer (ACS-Former) framework is proposed to maximize segmentation performance with limited annotation resources. The two-branch ACS-Former includes (1) a feature pyramid transformer (FPT) for multi-scale crack segmentation and (2) boundary difficulty-aware active learning (BDAL) to select informative images for labeling and incorporation into FPT training. Additionally, an adhesive climbing robot is proposed for image collection of hard-to-access components of large bridges. The on-site operational feasibility and practicability of the proposed ACS-Former and climbing robot are demonstrated by field experiments performed on in-service bridges, including the quantification of cracks narrower than 0.2 mm, as required by engineering codes.
混凝土桥梁裂缝的定量评估对于结构健康监测和数字孪生至关重要。然而,裂缝分割模型的训练严重依赖标注资源,其分割能力往往无法满足实际应用中遇到的细裂缝边界定位的准确性。本文提出了一种主动学习集成裂缝分割转换器(ACS-Former)框架,以在有限的标注资源下最大限度地提高分割性能。双分支 ACS-Former 包括:(1) 用于多尺度裂缝分割的特征金字塔转换器 (FPT);(2) 边界难度感知主动学习 (BDAL),用于选择信息图像进行标注并纳入 FPT 训练。此外,还提出了一种粘附式攀爬机器人,用于收集大型桥梁难以接近部件的图像。在使用中的桥梁上进行的现场实验证明了所建议的 ACS-Former 和攀爬机器人的现场操作可行性和实用性,包括按照工程规范的要求对小于 0.2 毫米的裂缝进行量化。
{"title":"Quantitative assessment of cracks in concrete structures using active-learning-integrated transformer and unmanned robotic platform","authors":"","doi":"10.1016/j.autcon.2024.105829","DOIUrl":"10.1016/j.autcon.2024.105829","url":null,"abstract":"<div><div>Quantitative assessment of cracks in concrete bridges is crucial for structural health monitoring and digital twin. However, the training of crack segmentation models relies heavily on annotation resources, and their segmentation capabilities are often unsatisfactory in terms of the accuracy of boundary location of thin cracks encountered in practice. In this paper, an active-learning-integrated crack segmentation transformer (ACS-Former) framework is proposed to maximize segmentation performance with limited annotation resources. The two-branch ACS-Former includes (1) a feature pyramid transformer (FPT) for multi-scale crack segmentation and (2) boundary difficulty-aware active learning (BDAL) to select informative images for labeling and incorporation into FPT training. Additionally, an adhesive climbing robot is proposed for image collection of hard-to-access components of large bridges. The on-site operational feasibility and practicability of the proposed ACS-Former and climbing robot are demonstrated by field experiments performed on in-service bridges, including the quantification of cracks narrower than 0.2 mm, as required by engineering codes.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142433452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Fast 3D site reconstruction using multichannel dynamic and static object separation 利用多通道动态和静态物体分离快速重建三维场地
IF 9.6 1区 工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY Pub Date : 2024-10-14 DOI: 10.1016/j.autcon.2024.105807
Three-dimensional (3D) models, characterized by their visualization, accuracy, and interactive information presentation, effectively facilitate collaboration and optimize management throughout the construction process. However, existing 3D reconstruction methods frequently fail to simultaneously satisfy the requirements for onsite applicability and fast performance. To address this challenge, this paper proposes a monocular camera-based 3D reconstruction method designed for onsite applicability and introduces dynamic–static separation to reduce the computational burden for faster processing. This approach enables the preestablishment of 3D models for static and dynamic objects. The positioning, pose, and orientation information of objects can be quickly integrated from multiple channels for fast 3D site reconstruction. Experimental results demonstrate that target objects can be identified across multiple channels and quickly integrated into 3D models. This paper offers both theoretical and practical contributions by enabling 3D reconstruction of construction sites using monocular cameras, which enhances project safety management and supports the implementation of digital twins.
三维(3D)模型具有可视化、精确性和交互式信息展示的特点,可有效促进整个施工过程中的协作和优化管理。然而,现有的三维重建方法往往无法同时满足现场适用性和快速性能的要求。为解决这一难题,本文提出了一种基于单目相机的三维重建方法,该方法专为现场应用而设计,并引入了动静分离技术,以减轻计算负担,加快处理速度。这种方法可以预先建立静态和动态物体的三维模型。物体的定位、姿态和方位信息可从多个通道快速集成,以实现快速三维现场重建。实验结果表明,目标物体可以通过多个通道识别,并快速集成到三维模型中。本文通过使用单目摄像机实现建筑工地的三维重建,提高了项目安全管理水平,并支持数字孪生的实施,从而为理论和实践做出了贡献。
{"title":"Fast 3D site reconstruction using multichannel dynamic and static object separation","authors":"","doi":"10.1016/j.autcon.2024.105807","DOIUrl":"10.1016/j.autcon.2024.105807","url":null,"abstract":"<div><div>Three-dimensional (3D) models, characterized by their visualization, accuracy, and interactive information presentation, effectively facilitate collaboration and optimize management throughout the construction process. However, existing 3D reconstruction methods frequently fail to simultaneously satisfy the requirements for onsite applicability and fast performance. To address this challenge, this paper proposes a monocular camera-based 3D reconstruction method designed for onsite applicability and introduces dynamic–static separation to reduce the computational burden for faster processing. This approach enables the preestablishment of 3D models for static and dynamic objects. The positioning, pose, and orientation information of objects can be quickly integrated from multiple channels for fast 3D site reconstruction. Experimental results demonstrate that target objects can be identified across multiple channels and quickly integrated into 3D models. This paper offers both theoretical and practical contributions by enabling 3D reconstruction of construction sites using monocular cameras, which enhances project safety management and supports the implementation of digital twins.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":null,"pages":null},"PeriodicalIF":9.6,"publicationDate":"2024-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142433449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
期刊
Automation in Construction
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
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
×
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
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