首页 > , Vol. , Issue () : -
车载移动测量系统可采集高精度道路三维点云数据，为道路边界自动化提取提供了支撑。为解决车载激光点云中城市道路边界点云提取困难问题，本文引入局部二值模式（Local Binary Pattern，LBP），针对各类城市道路边界特征，设计了高度LBP、高程离散度LBP和空间形状LBP三种改进算子；构建多元LBP特征语义识别模型，实现了道路路缘石空间几何与分布特征的量化分析；最后通过道路方向约束进行聚类去噪，提取道路边界。对4种不同的城市路段点云进行实验，实验数据的提取完整率为92.0%，准确率为95.8%。结果表明，该方法可以准确地提取不同环境下的道路边界，具有较强的适应性。
Objective：The vehicle mobile measurement system can quickly collect high-precision road 3D point clouds which is important data for road boundary automatic extraction. In order to solve the problem of difficult inaccurate extraction of urban road boundary point clouds in vehicle-borne laser point clouds. Method：Aiming at the characteristics of various urban road boundaries, this paper makes full use of the three-dimensional shape, spatial geometry and distribution characteristics of curbs, and introduces local binary pattern (LBP) to extract curbs point clouds. Three improved operators are designed including height LBP, elevation dispersion LBP and spatial shape LBP are designed. A multi-LBP feature semantic recognition model is built. This model realizes the quantitative expression of curbs and pavements with LBP eigenvalues. Finally, the road boundaries point clouds is extracted by cluster and denoising with the constraint of the road direction. Result：The point clouds of four different urban sections are tested. The extraction completeness rate of the experimental data was 92.0%, and the accuracy rate was 95.8%. Conclusion：The results show that the method can accurately extract road boundaries in different environments and has strong adaptability.