Three-dimensional intersection fine road information acquisition method and device based on trajectory big data

Abstract

The invention provides a three-dimensional intersection fine road information acquisition method and device based on track big data. Comprising the steps of track height sequence change trend extraction by utilizing a symbol segmentation aggregation approximate representation method, track optimization considering a multi-element feature space similarity model, three-dimensional intersection plane structure generation based on an increment fusion method, ramp separation (convergence) point extraction by utilizing a two-section threshold method, semantic segmentation of a road horizontal section and a ramp section, and semantic segmentation of a ramp section. Semantic enhancement fused with priori knowledge, and geometric structure information acquisition of the three-dimensional intersection. According to the method, the cost for obtaining the fine road information of the urban three-dimensional intersection is reduced, the detection method is simple and easy to implement, and experimental results show that the accuracy rate of the plane geometric position of the three-dimensional intersection is 94.22%, the accuracy rate of the ramp separation (convergence) point position is 89.64%, the accuracy rate of the road trend is 95.89%, and the cross-layer accuracy rate is 96.66%.

Description

technical field [0001] The invention relates to the research fields of geographic information systems and intelligent transportation, and more specifically, to a method and device for acquiring fine road information at three-dimensional intersections based on trajectory big data. Background technique [0002] Urban road information is the basis of intelligent transportation related applications. Among them, the intersection is an important part of the urban traffic road network, and the acquisition of fine road information is very important. The spatial location, range and detailed turning information of the intersection are the key to the traffic road network data at all levels, and the description of its geometric and topological information varies with the level of detail of the road network. At present, the data sources for obtaining urban road network information include: image data, LiDAR point cloud data, and vehicle trajectory data. In recent years, with the popula...

AI Technical Summary

Problems solved by technology

However, the above studies did not conduct an in-depth analysis and construction of the complex three-dimensional intersection structure

Images