The Extraction of Street Curbs from Mobile Laser Scanning Data in Urban Areas

Abstract

The demand for mobile laser scanning in urban areas has grown in recent years. Mobile-based light detection and ranging (LiDAR) technology can be used to collect high-precision digital information on city roads and building façades. However, due to the small size of curbs, the information that can be used for curb detection is limited. Moreover, occlusion may cause the extraction method unable to correctly capture the curb area. This paper presents the development of an algorithm for extracting street curbs from mobile-based LiDAR point cloud data to support city managers in street deformation monitoring and urban street reconstruction. The proposed method extracts curbs in three complex scenarios: vegetation covering the curbs, curved street curbs, and occlusion curbs by vehicles, pedestrians. This paper combined both spatial information and geometric information, using the spatial attributes of the road boundary. It can adapt to different heights and different road boundary structures. Analyses of real study sites show the rationality and applicability of this method for obtaining accurate results in curb-based street extraction from mobile-based LiDAR data. The overall performance of road curb extraction is fully discussed, and the results are shown to be promising. Both the completeness and correctness of the extracted left and right road edges are greater than 98%.

Robust Approach for Urban Road Surface Extraction Using Mobile Laser Scanning 3d Point Clouds

“…We compare the proposed algorithm with a recently introduced method for street curb extraction in urban environments (Zhao et al, 2021). This algorithm performs in three steps.…”

“…The authors claimed that their algorithm is successful in the scenarios of vegetation covering curbs, curved curbs and occluded curbs. The reader is referred to Zhao et al (2021) for more details on the curb extraction algorithm.…”

“…Understanding details and getting precise information about a road surface and its attached components (e.g., road curb, divider and roadside way) is a fundamental task of many applications of 3D (three-dimensional) city analysis including urban path planning, corridor mapping, accessibility diagnosis, road asset management, monitoring and reconstruction, autonomous vehicle navigation, mobile robotics, intelligent transportation, and overall, for ensuring road safety (Sun et al, 2019;Wen et al, 2019;Feng et al, 2021;Xu et al, 2021;Zhao et al, 2021;Sholevar et al, 2022).…”

“…The objective of this paper, i.e., extraction of road surface and its related components (road pavement, curb, divider and roadside way) is highly challenging, because in addition to the above complexities, more difficulties occur, e.g., roads have varying spectral features due to the various kinds of material (e.g., asphalt, cement, concrete, and gravel) used, and are regularly covered by hindrances (e.g., vehicles, bush and dense vegetation) present in the scenes. Moreover, these structures are intricate in nature, due to the absence of clearly defined edges, corners, while differences in widths, varying heights of curbs, and the large and inhomogeneous changes of curvature may confuse the extraction of detailed information of various road components (Cira et al, 2020;Romero et al, 2021;Zhao et al, 2021).…”

“…Many methods are available in literature for road surface and related components extraction and analysis (Kumar et al, 2013;Rodriguez-Cuenca et al, 2015;Bai et al, 2021;Zhao et al, 2021). Bottelier et al (2005) developed an approach for ground filtering of echo sounding data.…”