无人机三维激光扫描在地表沉降的监测研究

Monitoring of surface subsidence by UAV 3D laser scanning

由于矿产开发以及地下水过度消耗等原因,地表沉降乃至塌陷等现象近年来层出不穷,对高危区域的地表变化情况进行周期性监测的需求已显得十分迫切。本文针对地表沉降监测的特点及需求,为进一步提高监测的精度、效率和覆盖范围,提出了基于无人机平台的三维激光扫描技术,并采用小波神经网络方法对点云数据进行非线性逼近处理,使得曲面拟合的均方根误差控制在毫米级,进而提高建模精度。再通过对不同时期的地表三维模型进行配比,就可以提取变形特征和数据。实验证明,该方法的测量误差控制在7 mm以内,且监测效率和覆盖范围较传统手段有明显提升,在工程中具有一定的推广价值。

Due to mineral development and excessive consumption of groundwater,surface subsidence and even collapse have emerged one after another in recent years.The demand for periodic monitoring of surface changes in high-risk areas has become very urgent.According to the characteristics and requirements of surface settlement monitoring,in order to further improve the accuracy,efficiency and coverage of monitoring,this paper proposed a three-dimensional laser scanning technology based on UAV platform,used wavelet neural network method to process data,establishes three-dimensional models of surface in different periods,and extracted deformation features and data through model matching.Experiments showed that the measurement error of this method was controlled within a reasonable range,and the monitoring efficiency and coverage were significantly improved compared with traditional methods.It has a certain popularization value in engineering.