Automated recognition of 3D CAD model objects in laser scans and calculation of as-built dimensions for dimensional compliance control in construction

Frédéric Bosché

    Research output: Contribution to journalArticle

    Abstract

    The construction industry lacks solutions for accurately, comprehensively and efficiently tracking the three-dimensional (3D) status of buildings under construction. Such information is however critical to the successful management of construction projects: It supports fundamental activities such as progress tracking and construction dimensional quality control. In this paper, a new approach for automated recognition of project 3D Computer-Aided Design (CAD) model objects in large laser scans is presented, with significant improvements compared to the one originally proposed in Bosche et al. (in press) [11]. A more robust point matching method is used and registration quality is improved with the implementation of an Iterative Closest Point (ICP)-based fine registration step.

    Once the optimal registration of the project's CAD model with a site scan is obtained, a similar ICP-based registration algorithm is proposed to calculate the as-built poses of the CAD model objects. These as-built poses are then used for automatically controlling the compliance of the project with respect to corresponding dimensional tolerances.

    Experimental results are presented with data obtained from the erection of an industrial building's steel structure. They demonstrate the performance in real field conditions of the model registration and object recognition algorithms, and show the potential of the proposed approach for as-built dimension calculation and control. (C) 2009 Elsevier Ltd. All rights reserved.

    Original languageEnglish
    Pages (from-to)107-118
    Number of pages12
    JournalAdvanced Engineering Informatics
    Volume24
    Issue number1
    DOIs
    Publication statusPublished - Jan 2010

    Keywords

    • RANGE IMAGES
    • REGISTRATION
    • INSPECTION
    • VISION
    • SITES

    Cite this

    @article{266f851f026b48668adcfb2c60ac6f31,
    title = "Automated recognition of 3D CAD model objects in laser scans and calculation of as-built dimensions for dimensional compliance control in construction",
    abstract = "The construction industry lacks solutions for accurately, comprehensively and efficiently tracking the three-dimensional (3D) status of buildings under construction. Such information is however critical to the successful management of construction projects: It supports fundamental activities such as progress tracking and construction dimensional quality control. In this paper, a new approach for automated recognition of project 3D Computer-Aided Design (CAD) model objects in large laser scans is presented, with significant improvements compared to the one originally proposed in Bosche et al. (in press) [11]. A more robust point matching method is used and registration quality is improved with the implementation of an Iterative Closest Point (ICP)-based fine registration step.Once the optimal registration of the project's CAD model with a site scan is obtained, a similar ICP-based registration algorithm is proposed to calculate the as-built poses of the CAD model objects. These as-built poses are then used for automatically controlling the compliance of the project with respect to corresponding dimensional tolerances.Experimental results are presented with data obtained from the erection of an industrial building's steel structure. They demonstrate the performance in real field conditions of the model registration and object recognition algorithms, and show the potential of the proposed approach for as-built dimension calculation and control. (C) 2009 Elsevier Ltd. All rights reserved.",
    keywords = "RANGE IMAGES, REGISTRATION, INSPECTION, VISION, SITES",
    author = "Fr{\'e}d{\'e}ric Bosch{\'e}",
    year = "2010",
    month = "1",
    doi = "10.1016/j.aei.2009.08.006",
    language = "English",
    volume = "24",
    pages = "107--118",
    journal = "Advanced Engineering Informatics",
    issn = "1474-0346",
    publisher = "Elsevier Limited",
    number = "1",

    }

    TY - JOUR

    T1 - Automated recognition of 3D CAD model objects in laser scans and calculation of as-built dimensions for dimensional compliance control in construction

    AU - Bosché, Frédéric

    PY - 2010/1

    Y1 - 2010/1

    N2 - The construction industry lacks solutions for accurately, comprehensively and efficiently tracking the three-dimensional (3D) status of buildings under construction. Such information is however critical to the successful management of construction projects: It supports fundamental activities such as progress tracking and construction dimensional quality control. In this paper, a new approach for automated recognition of project 3D Computer-Aided Design (CAD) model objects in large laser scans is presented, with significant improvements compared to the one originally proposed in Bosche et al. (in press) [11]. A more robust point matching method is used and registration quality is improved with the implementation of an Iterative Closest Point (ICP)-based fine registration step.Once the optimal registration of the project's CAD model with a site scan is obtained, a similar ICP-based registration algorithm is proposed to calculate the as-built poses of the CAD model objects. These as-built poses are then used for automatically controlling the compliance of the project with respect to corresponding dimensional tolerances.Experimental results are presented with data obtained from the erection of an industrial building's steel structure. They demonstrate the performance in real field conditions of the model registration and object recognition algorithms, and show the potential of the proposed approach for as-built dimension calculation and control. (C) 2009 Elsevier Ltd. All rights reserved.

    AB - The construction industry lacks solutions for accurately, comprehensively and efficiently tracking the three-dimensional (3D) status of buildings under construction. Such information is however critical to the successful management of construction projects: It supports fundamental activities such as progress tracking and construction dimensional quality control. In this paper, a new approach for automated recognition of project 3D Computer-Aided Design (CAD) model objects in large laser scans is presented, with significant improvements compared to the one originally proposed in Bosche et al. (in press) [11]. A more robust point matching method is used and registration quality is improved with the implementation of an Iterative Closest Point (ICP)-based fine registration step.Once the optimal registration of the project's CAD model with a site scan is obtained, a similar ICP-based registration algorithm is proposed to calculate the as-built poses of the CAD model objects. These as-built poses are then used for automatically controlling the compliance of the project with respect to corresponding dimensional tolerances.Experimental results are presented with data obtained from the erection of an industrial building's steel structure. They demonstrate the performance in real field conditions of the model registration and object recognition algorithms, and show the potential of the proposed approach for as-built dimension calculation and control. (C) 2009 Elsevier Ltd. All rights reserved.

    KW - RANGE IMAGES

    KW - REGISTRATION

    KW - INSPECTION

    KW - VISION

    KW - SITES

    U2 - 10.1016/j.aei.2009.08.006

    DO - 10.1016/j.aei.2009.08.006

    M3 - Article

    VL - 24

    SP - 107

    EP - 118

    JO - Advanced Engineering Informatics

    JF - Advanced Engineering Informatics

    SN - 1474-0346

    IS - 1

    ER -