激光点云实现探地雷达图像地形校正的研究

摘要

探地雷达在地形起伏条件下采集数据时，必须考虑因地形变化产生的图像畸变和精确定位问题。本文采用车载激光扫描系统可快速获取研究区高精度的地形数据，通过探地雷达测量线上均匀分布的离散点，实现测量区地形数据与探地雷达图像之间的同步。根据时间移位原理和线性差值方法，选择某一标准的水平基准面作为参考平面，将探地雷达各道数据的双程传播时间进行校正，从而实现探地雷达图像的地形校正。校正后图像跟校正前相比，不仅显示出复杂地形下介质真实分布形态，而且有助于探地图像的解译和地下目标物及介质层的精确定位。%When ground penetrating radar data were collected under the undulating terrain,the image distortion and precise position caused by terrain fluctuation must be considered.A new method using the color point cloud data to compensate for elevation fluctuations is discussed.The high-precision three-dimensional color point cloud data can be quickly obtained by a vehicle-borne laser scanning system,and the synchronization of the GPR data and the terrain da-ta is achieved by the distributed markers in the measuring line.According to the time shift and linear interpolation,a flat datum level is selected as the reference,the two-way travel time of GPR data is corrected to implement topographic correction.At last,compared with the profile without topographic correction,the results of topographic correction could restore the true underground distribution under complex terrain and get a better profile interpretation and precise posi-tion of the objects and layers of the underground.