Three-Dimensional LiDAR Data Classifying to Extract Road Point in Urban Area

Abstract

The Light Detection and Ranging (LiDAR) system is one of the best ways to accurately and effectively gather 3-D terrain information. However, it is complicated to process the LiDAR cloud data due to its irregularity and large number of collected data points. This letter proposes a novel method to automatically extract urban road network from 3-D LiDAR data. This method uses height and reflectance of LiDAR data, and clustered road point information. Geometric information of general roads is also applied to correctly extract road points group. The proposed method has been tested on various urban areas which contain complicated road networks. The results demonstrate that the integration of height, reflectance, and geometric information of roads is a crucial factor that distinguishes the proposed method in its ability to reliably and correctly classify road points.

Urban Road Detection in Airbone Laser Scanning Point Cloud Using Random Forest Algorithm

“…To achieve this goal, pre-established constraints and filtering of the data transformed to the DSM are mostly used (e.g. Clode et al, 2004;Choi et al, 2008).…”

“…The authors combine approaches based on clustering using similar height and pulse intensity with additional constraints for height differences. Although the results are good, as authors point, the method is characterized by a high level of the false positive rate, especially on roofs of big, flat, and relatively low buildings (Choi et al, 2008).…”

“…Different approaches have been proposed for extracting of a road network from ALS data (e.g. Hu et al, 2004, Clode et al, 2004, Choi et al, 2008, Ferraza et al, 2016. The approaches that rely solely on ALS data base on the creating of a digital surface model (DSM) or a digital terrain model (DTM) and then performing transformations and filtering on the reduced dataset.…”

“…In order to filter the point cloud, Clode et al (2004) use only intensity and height differences in DSM. Another approach is proposed by Choi et al (2008). The authors combine approaches based on clustering using similar height and pulse intensity with additional constraints for height differences.…”

Automatic Extraction and Topology Reconstruction of Urban Viaducts From Lidar Data

“…It turns out to be effective to take the advantages from different data sources. To detect road just by the point cloud data, which relies on a single kind of sensors and can avoid the complexity of merging different data sources, is also proved acceptable and has achieved significant improvement (Choi and et al, 2008. Hu and et al, 2014.…”