Date of Award
Spring 4-16-2020
Semester of Degree
May
Document Type
Open Access Thesis
Degree Name
M.S. in Forest and Natural Resources Management
Department
Forest and Natural Resources Management
Major Professor
PAUL CROVELLA
Steering Committee Member
Dr. Mohamad Razkenari
Steering Committee Member
Dr. Endong Wang
Abstract
Existing buildings account for 40% of global energy consumption, and two-thirds of them will be still be operational in 2050. As most of these buildings lack the needed documentation for energy upgrades, it is essential to understand and represent the current conditions of their envelopes and mechanical systems. This project proposed a skeleton-based application for reconstructing and classifying pipes in existing buildings using point clouds from laser scanners and thermal images for Building Information Modeling (BIM) applications. MATLAB and Dynamo were used to process and model this information in Revit. Initial results indicate that the application is robust to identifying pipes and connections, and that thermal images can be used to create sematic-rich models. These results can contribute to improving the capabilities of some of the commercially available software for pipe reconstruction in BIM and to expediting the digital reconstruction processes in existing buildings.
Recommended Citation
BORGES RODRIGUES, PATRICK, "Point to Pipe: Automatic Reconstruction and Classification of Pipes Using Lasergrammetry and Thermogrammetry for Building Information Modeling (BIM)" (2020). Dissertations and Theses. 157.
https://digitalcommons.esf.edu/etds/157