SEGMENTATION AND 3D MODELING METHODS FOR STRUCTURAL ANALYSIS BY MACHINE LEARNING ALGORITHMS ON POINT CLOUDS OF TRADITIONAL WOODEN BUILDINGS

The purpose of this study is to develop a system to automatically generate a structural analysis model from a laser-scanned point cloud of a traditional wooden building. To achieve this, it is necessary to develop a segmentation method that extracts only the point cloud of structural members from the point cloud of the entire building without manual selection. As a solution to this problem, we developed a segmentation method using an unsupervised learning algorithm. The application of this algorithm automates a large part of the process involved in segmentation and reduces the amount of effort required.