Abstract
Building outlines are needed for various applications like urban planning, 3D city modelling and updating cadaster. Their automatic reconstruction, e.g. from airborne laser scanning data, as regularized shapes is therefore of high relevance. Today’s airborne laser scanning technology can produce dense 3D point clouds with high accuracy, which makes it an eligible data source to reconstruct 2D building outlines or even 3D building models. In this paper, we propose an automatic building outline extraction and regularization method that implements a trade-off between enforcing strict shape restriction and allowing flexible angles using an energy minimization approach. The proposed procedure can be summarized for each building as follows: (1) an initial building outline is created from a given set of building points with the alpha shape algorithm; (2) a Hough transform is used to determine the main directions of the building and to extract line segments which are oriented accordingly; (3) the alpha shape boundary points are then repositioned to both follow these segments, but also to respect their original location, favoring long line segments and certain angles. The energy function that guides this trade-off is evaluated with the Viterbi algorithm.
Highlights
Building outlines provide substantial information for the urban environment and are needed to map urban variation and change
This paper focuses on the regularization of extracted building points from a LIDAR point cloud
A good introduction to previous research in the field of regularization of building outlines is given by Jwa et al (2008)
Summary
Building outlines provide substantial information for the urban environment and are needed to map urban variation and change. Extracting building outlines or other urban features to keep data sets up to date is time and cost consuming. Airborne images and laser scanning have been a major data source for building outline extraction. Today’s airborne laser scanning technology can produce dense 3D point clouds with high accuracy, which makes it an eligible data source to reconstruct 2D building outlines or even 3D building models. This paper focuses on the regularization of extracted building points from a LIDAR point cloud. The proposed approach enables a reconstruction of building outlines with more than one main orientation and different angles by utilizing different methods to generate input for the energy function.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.