Research On Geochemical Vector Data Interpolation Algorithm Based On Moving Weighted Average Thin Plate Spline

With the development and maturity of spatial information technology,the related technologies of spatial data processing are also developing rapidly,and the thin plate spline algorithm also has a very important position in the field of geochemical vector data interpolation.In order to ensure the smoothness of the interpolation effect,the traditional thin plate spline interpolation algorithm will smooth the local high-value and low-value areas in the actual interpolation process,which will affect the accuracy of the interpolation result to a certain extent.This thesis aims at the problems and deficiencies of the thin plate spline interpolation algorithm in the field of geochemical vector data interpolation,through in-depth research and analysis of different geochemical interpolation methods and application examples,the existing thin plate spline interpolation algorithm was improved,so as to improve the accuracy of the interpolation results.Considering that there may be gross errors in the geochemical vector point data after manual collection and input,a double gross error discrimination method is proposed,which means that the Dixon discrimination method is used to find out the possible errors in the original geochemical data,and then use the Grubbs discriminant method to determine whether they are erroneous data.The data that is judged to be wrong by the double gross error discrimination method is eliminated to ensure that the parameters are more accurately estimated in the construction process of the basis function of the subsequent interpolation method.The experimental results show that the accuracy of the method for identifying data is improved by 0.4%,which effectively avoids the negative impact of gross error data on the interpolation accuracy;In order to solve the problem that the thin plate spline interpolation method smoothes the local high and low values for the smoothness of the interpolated raster data,which reduces the interpolation accuracy,a moving weighted average thin plate spline interpolation algorithm is proposed by combining the weighted average method,Shepard weighted model and thin plate spline interpolation algorithm.The experimental results show that the improved interpolation accuracy is increased by 13.795%,which significantly improves the interpolation accuracy of geochemical vector data;Finally,a prototype system of geochemical vector data interpolation with high versatility and accuracy based on the moving weighted average thin plate spline algorithm is designed and implemented,and an application example of geochemical vector data interpolation is constructed by taking a certain map of the Horqin Right Wing Middle Banner in Northeast China as the research object.