Automatic Recognition of Common Structural Elements from Point Clouds for Automated Progress Monitoring and Dimensional Quality Control in Reinforced Concrete Construction

Abstract

This manuscript provides a robust framework for the extraction of common structural components, such as columns, from terrestrial laser scanning point clouds acquired at regular rectangular concrete construction projects. The proposed framework utilizes geometric primitive as well as relationship-based reasoning between objects to semantically label point clouds. The framework then compares the extracted objects to the planned building information model (BIM) to automatically identify the as-built schedule and dimensional discrepancies. A novel method was also developed to remove redundant points of a newly acquired scan to detect changes between consecutive scans independent of the planned BIM. Five sets of point cloud data were acquired from the same construction site at different time intervals to assess the effectiveness of the proposed framework. In all datasets, the framework successfully extracted 132 out of 133 columns and achieved an accuracy of 98.79% for removing redundant surfaces. The framework successfully determined the progress of concrete work at each epoch in both activity and project levels through earned value analysis. It was also shown that the dimensions of 127 out of the 132 columns and all the slabs complied with those in the planned BIM.

Virtual Disassembling of Historical Edifices: Experiments and Assessments of an Automatic Approach for Classifying Multi-Scalar Point Clouds into Architectural Elements

“…A very general division of point cloud segmentation and classification is given in [24], in which the existing algorithms are divided into either the use of geometric axioms and mathematical functions, or the use of machine learning techniques. This division is concurrent with ideas presented by [25], in which the former is mentioned as the use of geometrical, spatial, and contextual constraints. The authors in [26] mentioned a distinction between heuristic and machine learning techniques.…”

“…While the appeal of machine learning is strong for performing point cloud processing in the case of complex geometries as encountered in heritage objects, the main bottleneck remains the generation of the training dataset [25]. In this paper, an algorithmic approach is considered in order to provide a fast result which may eventually be used to help generate training data for future machine learning techniques.…”

“…Algorithmic segmentation uses mathematical rules and functions as constraints during the segmentation (and possibly also classification). These rules may range from simple rules (e.g., "floors are flat and located below each storey" or "pillars are cylinders") [25] to the implementation of ontological relations [41][42][43].…”

“…In light of the largely manual operation in 3D modeling [25], its automation has therefore been a very interesting subject of research in recent years. The modeling of planar surfaces or façades has been studied in many works [47,48].…”

From Point Cloud Data to Building Information Modelling: An Automatic Parametric Workflow for Heritage

“…The reason for this is that certain BIM platforms, e.g., ArchiCAD®or Autodesk Revit® [34], enable the insertion of 2D drawings in PDF format. There are other works that addressed mesh or point cloud conversion algorithms to improve 3D elements in newly constructed buildings [35][36][37][38] and heritage buildings [8,9,27,39]. produced parametric objects with geometrical alterations and further information under the HBIM approach.…”