Efficient Vertical Object Detection in Large High-Quality Point Clouds of Construction Sites

Miguel Vega1, Alexander Braun1, Heiko Bauer2, Florian Noichl1, André Borrmann1
1 Chair of Computational Modeling and Simulation, Technical University of Munich, Germany
2 FARO EUROPE GmbH & Co.KG., Korntal-Münchingen, Germany


DOI: 10.35490/EC3.2021.156
Abstract: Even when adherence to the project schedule is a critical performance metric, still 53% of construction projects exhibit schedule delays. To contribute to efficient construction progress monitoring, a method is proposed to detect temporary objects in scans of construction sites. The proposed workflow includes: image processing, computer vision, and deep learning techniques. The method was tested on three real scans and with three object categories (cranes, scaffolds, and formwork). It achieved average rates above 88% for precision and recall and outstanding computational performance (1s to process 10^5points). These metrics demonstrate the method’s capability to segment point clouds of construction sites.
Keywords: Temporary Objects, Point cloud segmentation, Construction monitoring, TLS, Large point cloud
Pages: 148 - 157
Paper:
Contribution_156_final

Presentation Video: