Automatic filtering and 2D modeling of airborne laser scanning building point cloud

Abstract

This article suggests a new approach to automatic building footprint modeling using exclusively airborne LiDAR data. The first part of the suggested approach is the filtering of the building point cloud using the bias of the Z‐coordinate histogram. This operation aims to detect the points of roof class from the building point cloud. Hence, eight rules for histogram interpretation are suggested. The second part of the suggested approach is the roof modeling algorithm. It starts by detecting the roof planes and calculating their adjacency matrix. Hence, the roof plane boundaries are classified into four categories: (1) outer boundary; (2) inner plane boundaries; (3) roof detail boundaries; and (4) boundaries related to the missing planes. Finally, the junction relationships of roof plane boundaries are analyzed for detecting the roof vertices. With regard to the resulting accuracy quantification, the average values of the correctness and the completeness indices are employed in both approaches. In the filtering algorithm, their values are respectively equal to 97.5 and 98.6%, whereas they are equal to 94.0 and 94.0% in the modeling approach. These results reflect the high efficacy of the suggested approach.

Modeling Multi-Rotunda Buildings at LoD3 Level from LiDAR Data

“…Three-dimensional city models are often developed based on light detection and ranging (LiDAR) data, which are collected with the use of aerial and terrestrial remote sensing techniques [6,7]. The process of building modeling at various levels of detail, from LoD0 to LoD2, has been extensively investigated [8][9][10][11][12][13][14][15][16][17]. New approaches to modeling buildings are being proposed based on the density of point clouds [18], normal vectors on minimal subsets of neighboring LiDAR points to determine characteristic points in roof creases [15], shape descriptors, and cubes that divide the point cloud into roof surface segments [19].…”

“…However, even sophisticated techniques will not be able to handle some intrinsic modeling problems [20,21]. The density of point clouds acquired during airborne scanning of urban areas differs sometimes between roofs and walls, and the presence of outliers and noisy data can lead to errors in the process of generating point clouds and incorporating clouds into the reference system [17,22,23].…”

Hole Filling of Single Building Point Cloud Considering Local Similarity among Floors

“…The oblique photogrammetry or airborne LiDAR is the main way to collect threedimensional (3D) data of large-scale urban scenes [1,2]. However, the 3D model of an urban scene constructed via oblique photogrammetry or airborne LiDAR technology comprises a whole and a set of many objects [3]. Although this kind of reconstructed 3D model can satisfy the scene visualization requirements, it is difficult to perform recognition and attribute assignment to target objects.…”

Random Forest Machine Learning Technique for Automatic Vegetation Detection and Modelling in LiDAR Data

“…Furthermore, the studied zone may contain natural item classes such as vegetation, terrain, rivers, and lakes. In order to model the project area, its LiDAR point cloud has to be classified according to the main classes [2]. Once the classification is achieved successfully, the next step is to model each class aside.…”