复杂采空区激光扫描异常点云过滤研究

Filtering of Abnormal Point Cloud from Cavity Laser Scanner

针对复杂采空区激光探测点云数据处理过程中传统过滤方法无法过滤所有异常点的缺陷,综合运用弦高比和周长比判据及Open GL编程实现了异常点云拾取删除全面过滤。首先,在分析激光扫描轨迹线点云数据拓扑关系及异常点生成原因的基础上,综合运用弦高比和周长比判据对点云数据初步过滤;然后运用Open GL编程在显示界面上直接拾取删除异常点,实现了对点云数据二次精确全面过滤。工程应用实例表明,该方法不仅有效保留了采空区边界形态的完整性,也为矿山复杂采空区激光探测点云数据过滤提供了新思路,工程实用价值较高。

Due to the problem of abnormal points failed to be filtered with the traditional method during processing point data of underground complex cavity obtained by laser scanning,studies on complete filtering as well as the theory and method were stated. Firstly,based on the analysis of the topological relation of point cloud data on the laser sounding tracks and reason for abnormal point occurrence,preliminary filtration was conducted with the traditional filtering method for point cloud including both chord height ratio and perimeter ratio. Then,programming with Open GL was used to select and delete those abnormal points that were failed to be filtered out in the previous filtration,leading to an accurate and complete filtration in two steps. The following practical application in engineering project verified that such method can not only effectively retain the boundary integrity of the gob,but also as a new idea for filtering laser scanning point cloud,lay the foundation for the subsequent three-dimensional modeling and application.