论文部分内容阅读
在机场跑道异物检测和公路路面病害勘测等应用领域,快速高效获得大面积路面三维形貌至关重要。在利用车载单线式激光扫描仪实施路面三维重建的应用中,车辆的颠簸会使得路面三维点云数据出现失真。为此,使用RANSAC直线估计算法估计出由扫描仪获得的每条扫描线的斜率,进而消除了路面三维点云的不规则起伏失真。实验结果表明,优化前由车辆颠簸导致的路面点云起伏幅度达0.6 m,优化后的路面点云起伏幅度控制在0.04 m以内。路面重建效果经优化后有较大改进,此外点云优化算法耗时约11 s。因此,文中的路面三维点云优化方法原理简单、实施简便、路面点云优化质量高且耗时少,可在车辆存在颠簸时保证路面三维重建效果,在需要快速高效获得大面积路面三维形貌的场合具有较好的应用前景。
In the airport runway detection of foreign bodies and road pavement disease surveying applications, fast and efficient access to large-scale three-dimensional road surface morphology is essential. In the application of vehicle-mounted single-line laser scanner to implement 3D reconstruction of road surface, vehicle bumpy will distort the 3D point cloud data of the road surface. To this end, the RANSAC line estimation algorithm is used to estimate the slope of each scan line obtained by the scanner, thus eliminating the irregular undulation distortion of the three-dimensional point cloud of the road surface. The experimental results show that the fluctuation amplitude of point cloud caused by vehicle bump before optimization is 0.6 m, and the optimized fluctuation range of point cloud is within 0.04 m. Pavement reconstruction effect has been greatly improved after optimization, in addition point cloud optimization algorithm takes about 11 s. Therefore, the optimization method of road surface 3D point cloud in this paper is simple in principle and easy to implement. The optimization of road surface point cloud is of high quality and less time consuming. It can ensure the three-dimensional reconstruction effect of road when it is bumpy. The occasion has a good application prospects.