DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

3D Building Reconstruction from Airborne LiDAR Point Clouds

Orthuber, Elisabeth (2014) 3D Building Reconstruction from Airborne LiDAR Point Clouds. Master's, TU München.

[img] PDF


This work presents a novel workflow for data-driven building reconstruction from point clouds acquired with aerial Light Detection and Ranging (LiDAR) sensors. The goal of this thesis is to produce 3D building models of high accuracy and level of detail, including roof superstructures such as dormers. Roof segments shall be connected to each other by intersection edges or step edges, represented by vertical walls. The workflow comprises the extraction of building point clouds from the LiDAR scene, roof segmentation, segment boundary creation, and 3D modeling. The workflow is tested and evaluated for two data sets, using the evaluation method and test data of the “ISPRS Test Project on Urban Classification and 3D Building Reconstruction”. LiDAR points of buildings are extracted from the scene using previously available 2D building boundary polygons. Nearby points from terrain and vegetation are removed using filtering procedures. For roof segmentation, a robust region growing technique is developed. A unique feature of the segmentation method is the growing of triangles of a Triangulated Irregular Network (TIN) instead of LiDAR points. This minimizes the gaps between segments, because LiDAR points at segment intersections can be assigned multiple segment labels. Additionally, robust adaptive thresholds are introduced as region growing criteria. These enable the region growing procedure to stop at weak edges, while also segmenting non-planar roof segments. Results show that the proposed segmentation outperforms other methods concerning undersegmentation, and that it recognizes even weak edges. Evaluation and an extensive analysis of the input parameters’ effects on the results have shown that the segmentation is very robust against LiDAR point cloud characteristics and segment shape. Segment boundaries are cretated by collapsing the convex hull of segment points. Point density variations in across-track and along-track directions are considered in the collapsing procedure. For building modeling, the 2.5D dual contouring approach of Zhou and Neumann [2010] is adapted to model complex roofs. After overlying a 2D grid to the segmented point cloud, vertices of the 3D building model are estimated for each grid cell by minimizing a Quadratic Error Function (QEF). Each QEF minimization results in a hyperpoint, which consists of one or more vertices of the building model at the same x-y-coordinates. This 2.5D-characteristic enables the connection of building vertices at step edges with vertical walls. In contrast to Zhou and Neumann [2010], the proposed method uses a detailed roof segmentation, where segments can be connected by step edges or intersection edges to each other. The main contribution of this work is the modification and weighting of the QEF, such that the number of hyperpoint vertices resulting from QEF minimization adapts to whether step edges or intersection edges shall be modeled. For enhancing model simplicity, the QEF solutions of all cells are merged by collapsing a quadtree. The quadtree collapsing stops, when the QEF residual exceeds a user-defined threshold. The final hyperpoint vertices are connected according to their adjacency in the quadtree to form 3D faces of the polygonal building model. Testing of the procedure has shown that the resulting building models are very detailed and watertight. Building superstructures can be represented accurately, and also non-flat roof segments can be modeled in detail. Each roof segment is represented by a triangulation of the building vertices, such that the building models precisely fit the input data, depending on the flexibly chosen input parameters. As the building models are composed of many polygon faces, subsequent regularization is recommended to further enhance model simplicity. Evaluation has shown that both proposed segmentation and reconstruction methods outperform other methods in important quality measures. The tested scenes show outstanding completeness and undersegmentation, and comparative planimetric accuracy compared to scenes produced with other reconstruction methods. An extensive analysis of the impact of input parameters on the results has shown that the procedure is very robust. The user can influence the level of detail of the builing models by choosing the input parameters. To the best knowledge of the author, the proposed reconstruction method is the first dual contouring approach for modeling complex roof height layers.

Item URL in elib:https://elib.dlr.de/92520/
Document Type:Thesis (Master's)
Title:3D Building Reconstruction from Airborne LiDAR Point Clouds
AuthorsInstitution or Email of AuthorsAuthors ORCID iD
Date:18 September 2014
Refereed publication:No
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:No
Number of Pages:75
Keywords:LiDAR 3D recosntruction building extraction
Institution:TU München
Department:Chair of Remote Sensing Technology
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Erdbeobachtung
DLR - Research theme (Project):R - Vorhaben hochauflösende Fernerkundungsverfahren
Location: Oberpfaffenhofen
Institutes and Institutions:Remote Sensing Technology Institute > Photogrammetry and Image Analysis
Deposited By: Avbelj, Janja
Deposited On:01 Dec 2014 10:08
Last Modified:31 Jul 2019 19:49

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.