University of Twente Student Theses
Scan-vs-BIM automated registration using columns segmented by deep learning for construction progress monitoring
Tsige, Girma Zewdie (2022) Scan-vs-BIM automated registration using columns segmented by deep learning for construction progress monitoring.
|
PDF
2MB |
| Abstract: | In construction automation applications, registration between 3D Building Information Modelling (BIM) and the point cloud obtained by remote sensing technology is critical for smart monitoring of construction progress. The existing AEC/FM software packages for 3D data registration implement coarse registration manually, which is less accurate and more time-consuming. In this work, we present a novel automated column-based coarse registration method for 3D point clouds with the 3D BIM model from building construction sites. The method is based on the extraction of columns from the as-built point cloud and the 3D BIM model. For the point cloud data, fully automated column extraction techniques are used by applying deep learning, whereas columns are extracted from the structural-semantic component for the BIM model. Then, we propose an automatic coarse registration method that is motivated by the Random Sample Consensus (RANSAC) algorithm.; we call it the ‘congruent column sets’ algorithm. Experiments were carried out on as-built point clouds acquired from the real building construction site using both terrestrial laser scan (TLS) and unmanned aerial vehicles (UAV) to validate the proposed method. The results show that our proposed column-based registration method achieved a rotation error of 0.02 degrees and RMSE of 0.12 meters for the TLS dataset and 0.03 degrees and 0.17meters for the UAV dataset. We conclude that our proposed approach contributes to automating the registration between the as-built point cloud and the as-planned BIM model to monitor the construction progress. |
| Item Type: | Essay (Master) |
| Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
| Subject: | 54 computer science, 56 civil engineering |
| Programme: | Geoinformation Science and Earth Observation MSc (75014) |
| Link to this item: | https://purl.utwente.nl/essays/92075 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page