Abstract
Abstract. Classification and segmentation of buildings from airborne lidar point clouds commonly involve point features calculated within a local neighborhood. The relative change of the features in the immediate surrounding of each point as well as the spatial relationships between neighboring points also need to be examined to account for spatial coherence. In this study we formulate the point labeling problem under a global graph-cut optimization solution. We construct the energy function through a graph representing a Markov Random Field (MRF). The solution to the labeling problem is obtained by finding the minimum-cut on this graph. We have employed this framework for three different labeling tasks on airborne lidar point clouds. Ground filtering, building classification, and roof-plane segmentation. As a follow-up study on our previous ground filtering work, this paper examines our building extraction approach on two airborne lidar datasets with different point densities containing approximately 930K points in one dataset and 750K points in the other. Test results for building vs. non-building point labeling show a 97.9% overall accuracy with a kappa value of 0.91 for the dataset with 1.18 pts/m2 average point density and a 96.8% accuracy with a kappa value of 0.90 for the dataset with 8.83 pts/m2 average point density. We can achieve 91.2% overall average accuracy in roof plane segmentation with respect to the reference segmentation of 20 building roofs involving 74 individual roof planes. In summary, the presented framework can successfully label points in airborne lidar point clouds with different characteristics for all three labeling problems we have introduced. It is robust to noise in the calculated features due to the use of global optimization. Furthermore, the framework achieves these results with a small training sample size.
Highlights
Simpler representations derived from airborne lidar point clouds usually provide more practical and manageable input for efficient analysis in many applications
The first dataset we used in this research is the lidar point cloud of part of Bloomington, Indiana of USA obtained from the Indiana Spatial Data Portal
The point coordinates are in NAD 1983 HARN horizontal datum and NAVD 88 vertical datum projected on the Indiana West State Plane Coordinate System
Summary
Simpler representations derived from airborne lidar point clouds usually provide more practical and manageable input for efficient analysis in many applications. Earlier research focuses more on handling lidar data as 2.5D by resampling the point clouds into raster grids Such approaches may be convenient for employing a wide range of wellestablished image processing algorithms, but loss of information is inevitable due to resampling. Niemeyer et al (2012) utilize CRF formulation for the supervised classification of lidar point clouds using loopy belief propagation (LBP) for inference In their later research, Niemeyer et al (2014) employ random forest (RF) classification to calculate the pairwise potentials of CRF model for labeling airborne point clouds. We propose a framework which takes advantage of the contextual formulation capabilities of MRFs coupled with powerful graph-cut optimization for semantically labeling point clouds with focus on building extraction. We introduce a new approach for utilizing the point features for the calculation of data and smoothness costs of 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 Annals 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.