Post-Processing Methods and Applications for Ship Block 3D Scanning Data Based on Open-Source Point Cloud Algorithms

Recently, in the shipbuilding industry, various smart quality management technologies integrating Fourth Industrial Revolution technologies have been actively researched, with particular emphasis on the use of 3D laser scanning technologies for the rapid acquisition of precise data. The raw 3D point cloud data generated by 3D laser scanners often contains incomplete information and noise due to various factors such as surface characteristics, measurement environment, and weather conditions. Consequently, post-processing of the data is essential to accurately extracting and analyzing the desired information. Accordingly, this study investigates post-processing methods and applications for the analysis of point cloud in ship blocks. Utilizing open-source algorithms, the research implements and evaluates the feasibility of 3D point cloud data processing and analysis techniques, including noise removal and plane recognition. Initially, the algorithms were applied to sample 3D point cloud data measured in a laboratory setting, and an examination of characteristics based on algorithm parameters confirmed their applicability. Based on these findings, a step-by-step assessment of each algorithm was conducted on point cloud data from ship blocks measured on-site. The results indicated the applicability of the algorithms, except for the DBSCAN algorithm, for the ship block point cloud data.