To facilitate the maintenance and operation process of existing bridges, there is an urgent need to digitalize the inspection, condition assessment, and evaluation of bridges. Part of this task is to automate the generation of a 3D model from point cloud data (PCD). Therefore, MAFIPOUR et al. (2021) proposed a new parametric model fitting method that can automatically generate arbitrary 3D models based on PCD. This thesis focuses on the reconstruction of the bridge deck. A parametric model is fitted to the point cloud using a metaheuristic algorithm to automatically reconstruct the bridge deck based on the PCD. The novelty of this method is to employ reverse engineering with parametric modeling to extract the value of parameters from point clouds. The described method was tested on six bridges with arbitrary deck shapes. The results show that an average mean absolute error (MAE) of 4.21 cm and an average processing time of 90.79 sec per bridge is achieved. Thus, the parametric model fitting-to-cloud algorithm provides a close estimation in a short time, plus important information about the dimension of a component is automatically extracted and used to generate a complex shaped bridge deck.     «

To facilitate the maintenance and operation process of existing bridges, there is an urgent need to digitalize the inspection, condition assessment, and evaluation of bridges. Part of this task is to automate the generation of a 3D model from point cloud data (PCD). Therefore, MAFIPOUR et al. (2021) proposed a new parametric model fitting method that can automatically generate arbitrary 3D models based on PCD. This thesis focuses on the reconstruction of the bridge deck. A parametric model is fi...     »