Towards Reconstructing 3D Buildings from ALS Data Based on Gestalt Laws

Abstract

3D building models are an essential data infrastructure for various applications in a smart city system, since they facilitate spatial queries, spatial analysis, and interactive visualization. Due to the highly complex nature of building structures, automatically reconstructing 3D buildings from point clouds remains a challenging task. In this paper, a Roof Attribute Graph (RAG) method is proposed to describe the decomposition and topological relations within a complicated roof structure. Furthermore, top-down decomposition and bottom-up refinement processes are proposed to reconstruct roof parts according to the Gestalt laws, generating a complete structural model with a hierarchical topological tree. Two LiDAR datasets from Guangdong (China) and Vaihingen (Germany) with different point densities were used in our study. Experimental results, including the assessment on Vaihingen standardized by the International Society for Photogrammetry and Remote Sensing (ISPRS), show that the proposed method can be used to model 3D building roofs with high quality results as demonstrated by the completeness and correctness metrics presented in this paper.

Constraint-based automated reconstruction of grape bunches from 3D range data for high-throughput phenotyping

“…Another application field is the reconstruction of complex objects in 3D data. One class of methods aims at the reconstruction of buildings, focusing on aspects like the reconstruction of the building from airborne LiDAR data (Hu et al, 2018, Wu et al, 2017a, a detailed facade reconstruction (Dehbi et al, 2016) or the derivation of the building model based on pointclouds generated indoors (Ochmann et al, 2016). Dehbi et al (2016) propose to derive geometrical, topological, and semantic constraints from examples to generate a better reconstruction.…”

“…Dehbi et al (2016) propose to derive geometrical, topological, and semantic constraints from examples to generate a better reconstruction. Hu et al (2018) formulate constraints as descriptions of decompositions and topological relations between primitives. In the field of indoor scene reconstructions scene databases with manually segmented and labeled mesh models are available and can be used to learn contextual relationships as constraints between objects (Chen et al, 2014).…”

Comparison of Software for Airborne Laser Scanning Data Processing in Smart City Applications

“…All type of laser scanning technologies (terrestrial, mobile, airborne) are a very popular and powerful tool for urban territory surveys to collect data during Smart City projects implementation (Balado et al, 2018;Hu et al, 2018;Li et al, 2018;Julin et al, 2018;Badenko et al, 2019;Yang and Lee, 2019). Airborne LIDAR (Light Identification Detection and Ranging) is one of the optimal modern surveying methods for urban environment (Jochem et al, 2012;Tomljenovic et al, 2015;Yan et al, 2015).…”

Evaluating the Possibility of Tree Species Classification With Dual-Wavelength Als Data

“…ALS technology provides the user with accurate geometric information about registered objects. Point clouds, which are derived from ALS, are mostly used for the generation of digital terrain models (DTMs), digital surface models (DSMs) and in 3D modelling (Hu et al, 2018). Another advantage of ALS is laser beam penetration through vegetation, which not only makes it possible to obtain data on the tree canopy but also within and under the tree.…”