Version 1
: Received: 29 June 2023 / Approved: 30 June 2023 / Online: 30 June 2023 (12:47:42 CEST)
How to cite:
Mansour, M.; Martens, J.; Blankenbach, J. Hierarchical SVM for Semantic Segmentation of 3D Point Clouds for Infrastructure Scenes. Preprints2023, 2023062255. https://doi.org/10.20944/preprints202306.2255.v1
Mansour, M.; Martens, J.; Blankenbach, J. Hierarchical SVM for Semantic Segmentation of 3D Point Clouds for Infrastructure Scenes. Preprints 2023, 2023062255. https://doi.org/10.20944/preprints202306.2255.v1
Mansour, M.; Martens, J.; Blankenbach, J. Hierarchical SVM for Semantic Segmentation of 3D Point Clouds for Infrastructure Scenes. Preprints2023, 2023062255. https://doi.org/10.20944/preprints202306.2255.v1
APA Style
Mansour, M., Martens, J., & Blankenbach, J. (2023). Hierarchical SVM for Semantic Segmentation of 3D Point Clouds for Infrastructure Scenes. Preprints. https://doi.org/10.20944/preprints202306.2255.v1
Chicago/Turabian Style
Mansour, M., Jan Martens and Jörg Blankenbach. 2023 "Hierarchical SVM for Semantic Segmentation of 3D Point Clouds for Infrastructure Scenes" Preprints. https://doi.org/10.20944/preprints202306.2255.v1
Abstract
Building Information Modelling (BIM) has gained significant relevance in civil engineering, particularly in the infrastructure sector. It offers the potential to increase efficiencies, optimize processes, and enhance sustainability throughout infrastructure projects’ life cycles. BIM consolidates information about asset components into a single model, enabling real-time updates, cost reduction, and improved consistency. Digital twins serve as valuable extensions of BIM in infrastructure, playing a crucial role during operation. By capturing real-time data from sensors and IoT devices, digital twins enabls continuous monitoring, proactive maintenance, efficient energy management, and informed decision-making for optimized operational efficiency. Advancements in 3D point cloud technologies, such as 3D laser scanning, have expanded BIM’s application in operation and maintenance. These technologies rapidly capture accurate and detailed three-dimensional information, prompting the exploration of automated alternatives to manual point cloud modeling. This paper demonstrates the application of supervised machine learning, specifically support vector machines, for analyzing and segmenting 3D point clouds, a crucial step in 3D modeling. Various approaches for semantic segmentation are introduced, investigated, and evaluated using diverse data sets. The results highlight the effectiveness of supervised machine learning techniques in achieving accurate segmentation of 3D point clouds.
Keywords
Building information modelling; machine learning; infrastructure; 3D laser scanning; semantic segmentation; support vector machine.
Subject
Engineering, Architecture, Building and Construction
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
