数字孪生基础数据底板DEM修正方法研究

Study on DEM Correction Method of Digital Twin Basic Data Baseboard

DEM是最重要的数字孪生基础数据之一,在水利规划设计以及工程建设等领域发挥了巨大的作用。然而各类DEM误差的存在不同程度地降低了分析与应用结果的可信度,特别是地形转DEM后的缺失值大大限制了后续基于DEM的分析使用。从栅格计算原理出发,通过Python调用ArcGIS相应的工具,对某区域的地形转DEM后接幅处出现的缺失值,采用该像元周边8个有值像元的均值来修正。结果表明:计算方法适用性强,结果可靠,计算速度快。研究区域内306个NoData像元的修正值和比较值的数据紧靠1∶1线,修正值和比较值的最大差值仅为3.55 m,中误差值0.75,远低于规范的误差限。通过Python调用Numpy,极大提高计算速度(本次计算用时10 s),而ArcGIS暂时没有该方法的直接修正工具。因此,提出的DEM修正方法可批量处理,不存在内存限制处理失败的情形,极大提高工作效率,对数字孪生基础数据处理有较强的应用价值。

DEM is one of the most important digital twin basic data and has played a huge role in water conservancy planning,design,and engineering construction.However,the existence of various DEM errors greatly reduces the credibility of analysis and application results.In particular,the missing value after terrain conversion to DEM greatly limits subsequent DEM-based analysis.Based on the principle of raster calculation,this paper calls the corresponding tool of ArcGIS by Python to correct the missing value at the overlapped area of a certain terrain converted to DEM by using the mean of eight valued pixels around the pixel.The results show that the calculation method is highly applicable,reliable,and fast.The data of the corrected and compared values of 306 NoData pixels in the study area are close to the 1∶1 line;the maximum difference between the corrected and compared values is only 3.55 m,and the medium error value is 0.75,which is far below the standard error limit.Calling Numpy through Python greatly improves the calculation speed(10 s),and ArcGIS currently has no tool for directly correcting this method.In contrast,the DEM correction method proposed in this paper can be processed in batch,without processing failures due to memory limit,which greatly improves the work efficiency and has a strong application value for processing digital twin basic data.