Road boundary extraction method from mobile laser scanning point clouds

With the rapid development of mobile laser scanning (MLS) technology, high-precisionthree-dimensional (3D) point cloud data has shown great potential in different fields such astopographic mapping, road asset management and smart city construction. 3D point cloud datacontains not only position information, but also the shape and attributes of the target, which isvery convenient for obtaining road information. Accurate extraction of the road boundary is abasic task for obtaining road infrastructural data, which can support the generation ofhigh-precision maps, vehicle navigation and autonomous driving. However, road boundaryextraction in urban environments is easily occluded such as vehicles and pedestrians on theroad, which leads to problems such as difficulty and incomplete extraction of MLS point cloudroad boundaries. To address this problem, this study proposes a road boundary extractionmethod that integrates pavement edge information, accurately considers the position of the roadboundary from two dimensions, eliminates false boundaries, and completes the missingboundary through the extracted boundary spatial relationship. First, grid elevation filtering isused to remove high-level non-ground points. Then the pavement edges and curb stone pointsare extracted from the preprocessed point cloud, and they are superimposed to remove falseboundary points to obtain accurate road boundaries. Finally, based on the spatial relationship ofroad boundaries, missing parts are detected and repaired to obtain complete road boundaries.Experimental results show that the accuracy on real road scenes exceeds 98%, the completenessrate is above 91%, and the extraction quality is above 90%, which verifies the effectiveness andaccuracy of this method.