Automatic recognition of a piping system from laser scanned points by eigenvalue analysis

摘要

In recent years, changes in plant equipment have been becoming more frequent, and as-built modeling of plants from large-scale laser scanned data is expected to make their rebuilding processes more efficient. The purpose of our research was to propose an algorithm which can automatically recognize piping systems from massive terrestrial laser scanned point clouds of plants. Point clouds of a piping system can be extracted based on eigenvalue analysis and using region-growing from the laser scanned points. Eigenvalue analysis of the point clouds and point normals then allows for the recognition of straight portions of pipes. Connecting parts can also be recognized from the connection relationship between pipe axes and their neighboring scanned point distributions.