Abstract
Aerial laser scanning or photogrammetric point clouds are often noisy at building boundaries. In order to produce regularized polygons from such noisy point clouds, this study proposes a hierarchical regularization method for the boundary points. Beginning with detected planar structures from raw point clouds, two stages of regularization are employed. In the first stage, the boundary points of an individual plane are consolidated locally by shifting them along their refined normal vector to resist noise, and then grouped into piecewise smooth segments. In the second stage, global regularities among different segments from different planes are softly enforced through a labeling process, in which the same label represents parallel or orthogonal segments. This is formulated as a Markov random field and solved efficiently via graph cut. The performance of the proposed method is evaluated for extracting 2D footprints and 3D polygons of buildings in metropolitan area. The results reveal that the proposed method is superior to the state-of-art methods both qualitatively and quantitatively in compactness. The simplified polygons could fit the original boundary points with an average residuals of 0.2 m, and in the meantime reduce up to 90% complexities of the edges. The satisfactory performances of the proposed method show a promising potential for 3D reconstruction of polygonal models from noisy point clouds.
Highlights
With the rapid developments in aerial laser scanning (ALS) and aerial oblique photogrammetry, 3D point clouds have become the primary datasets used in large-scale urban reconstruction [1,2]
Unlike other modified Manhattan-based regularization methods which need to pre-define the number of resulting orientations [18] or difference angle threshold for compulsive orthogonality [19], the proposed method casts the problem as a Markov random field (MRF) and allows multiple and arbitrary orientations with mutual regularities in the same optimization; and (3) we propose a strategy to produce simplified polygons with inter-part regularity in 3D space
Qualitative and quantitative comparisons of the proposed method with existing methods have revealed the effectiveness of the proposed method in handling highly noisy point boundaries and producing polygons with both satisfactory regularity and data fidelity
Summary
With the rapid developments in aerial laser scanning (ALS) and aerial oblique photogrammetry, 3D point clouds have become the primary datasets used in large-scale urban reconstruction [1,2]. In the boundary simplification process, traditional edge collapse-based methods are likely to eliminate sharp features, whereas regularization methods that adopt the Manhattan rule based on the dominant orientation tend to be too strict in many real-world applications, leading to large distortions when applied to polygons with multiple orientations. We extend our preliminary conference paper [20] in the following four directions: (1) improving the local stage regularization algorithm, allowing it to handle the zigzag effect more efficiently; (2) taking regional level relationships into considerations to handle boundary regularities between different planes or buildings; (3) conducting additional experiments on relatively low-density ALS data for building footprint generalization; and (4) supplementing more quantitative comparison with other representative methods.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
