Detection of doors in a voxel model, derived from a point cloud and its scanner trajectory, to improve the segmentation of the walkable space

B. R. Staats, A. A. Diakité, R. L. Voûte and S. Zlatanova

International Journal of Urban Sciences, 2019, vol. 23, issue 3, 369-390

Abstract: Generation of indoor networks for navigation will normally be done out of standard floor plans that are only 2D and is more often manual than automatic. These floor plans are drawn at a specific time and do not correspond to the reality, moreover some of those buildings were built already differently than designed. Then in due course the building will change both externally and internally. Also objects like furniture will be moved around in the building. If these changes are not updated in the map of the building, it becomes out of date and cannot be used for the creation of indoor navigable models anymore. To enable correct indoor navigation, we will need to have the current data of the indoor environment. This article concentrates on providing a new approach to generate up to date floor plans by using a mobile (and hand held) laser scanner in the fastest way. This device creates a point cloud and the corresponding trajectory at the same time. Because the mobile laser scanner device is operated by a walking human, the trajectory contains information about the surface the person is walking on. In this article, a method is explained for the detection of walkable spaces based on the analysis of the point cloud and its corresponding trajectory provided by the mobile laser scanner. Three steps will be used: voxelization, trajectory analysis and the identification of floor regions. Dynamic objects, doors, and furniture objects are also used to identify the surfaces which are available for navigation purposes. Three types of surfaces are considered: horizontal, slopes, and stairs.

Date: 2019
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/12265934.2018.1553685 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:rjusxx:v:23:y:2019:i:3:p:369-390

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/rjus20

DOI: 10.1080/12265934.2018.1553685

Access Statistics for this article

International Journal of Urban Sciences is currently edited by Dongjoo Park and Mack Joong Choi

More articles in International Journal of Urban Sciences from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().