Space Subdivision in Indoor Mobile Laser Scanning Point Clouds Based on Scanline Analysis

Abstract

Indoor space subdivision is an important aspect of scene analysis that provides essential information for many applications, such as indoor navigation and evacuation route planning. Until now, most proposed scene understanding algorithms have been based on whole point clouds, which has led to complicated operations, high computational loads and low processing speed. This paper presents novel methods to efficiently extract the location of openings (e.g., doors and windows) and to subdivide space by analyzing scanlines. An opening detection method is demonstrated that analyses the local geometric regularity in scanlines to refine the extracted opening. Moreover, a space subdivision method based on the extracted openings and the scanning system trajectory is described. Finally, the opening detection and space subdivision results are saved as point cloud labels which will be used for further investigations. The method has been tested on a real dataset collected by ZEB-REVO. The experimental results validate the completeness and correctness of the proposed method for different indoor environment and scanning paths.

Automatic 3-D Reconstruction of Indoor Environment With Mobile Laser Scanning Point Clouds

“…The method relied on the prior knowledge of scanning stations, which caused over segmentations for long corridors. Mura [13], [40] reconstructed individual rooms in 3D environments with arbitrary wall orientations and the results outperformed the 2.5D reconstruction results of other approaches [12], [14], [17], [36], [38]. However, the method was only applied to small-scale scenes.…”

“…Other state-of-the-art methods [12], [13], [39]- [41] can only be applied to TLS point clouds, resulting in limited usage in wider applications using low cost MLS systems. Some methods [3], [14], [16]- [19] are proposed to build multi-room models, depending on iterative heuristics or subsequent merging steps, using MLS point clouds. Other methods [6], [11], [42] for large scene reconstruction can only express the building's geometry and structural information, but are in lack of space recognition and semantic information.…”

“…However, most of the methods are limited to TLS point clouds that are inefficient to acquire. Currently, more and more work [3], [14]- [19] have constructed indoor models using MLS systems, as they offer not only a less occluded point cloud but also a trajectory followed by the system during the acquisition process [18]. Nevertheless, the reconstruction methodologies based on MLS are highly dependent on data quality, integrity and are mostly applied to simple scenarios.…”

Space Subdivision of Indoor Mobile Laser Scanning Data Based on the Scanner Trajectory

“…Finally, the individual rooms are segmented and extruded to construct the model. Zheng et al [18] proposed an approach for subdividing the indoor spaces by analyzing the individual scanlines of the mobile laser scanners. The scanlines are used to detect openings such as doors and windows.…”

“…Also, it refers to semantic place labeling in the robotics domain [17]. Several studies automated the process of room segmentation for domain-specific applications [18][19][20][21][22]. In the same context, the trajectory of the mobile laser scanner can reveal the topological relations between the spaces inside the building.…”