Research On Geological Surface Reconstruction Method With Non-parallel Polygons

Taking the seismic data and the logging data as the basic data source,in view of the lack of research on surface reconstruction of geological bodies based on non-parallel 3D polygons,smooth and accurate geologic surface is rebuilt based on a point cloud constructed from an ordered discrete point on a polygon.The content of this thesis is divided into the following aspects:Firstly,in the curve fitting and point cloud acquisition stage,in order to make the contours of the space continuous smooth,a continuous segment straight-line & circular interpolation algorithm is designed and implemented,which constructs a set of smooth and ordered boundary points.In order to construct point cloud,resampling method was used with appropriate resampling rates to select.Secondly,in the surface reconstruction phase,the shield Poisson Surface Reconstruction was used with following aspects: normal vector information preprocessing,global problem discretization,solving Poisson equation,isometric surface extraction and other steps.Finally,in the post-processing stage,mainly using Laplacian surface smoothing and triangular mesh subdivision,the smoothness of the surface is improved by Laplacian Smooth,and the local accuracy is increased by triangular mesh subdivision.These methods is nested together to make the surface reconstruction results as much as possible towards the original geological body surface.