3D Near-surface Modeling And Visualization Based On Spatial Interpolation

The near surface is a special structure that affects the accuracy of seismic exploration imaging,and it is the focus of geologists' research.Over the years,geologists have studied and improved near-surface anomaly correction and geophysical near-surface velocity modeling methods,but few studies have been conducted on near-surface 3D modeling and visualization.Near the surface of the real rock layer above the ground,the medium composition is complex and the propagation velocity varies greatly.Usually,it is not a strict layered structure.Therefore,it is difficult to construct a layered model of near-surface by constructing triangular facets.In this paper,a near-surface attribute modeling method based on spatial interpolation is proposed.First,the geological block is divided and filled with attribute data.The near-surface attribute model can be constructed and drawn through attribute data body interpolation and regular data field visualization.Spatial interpolation is a data pre-processed method that intervenes sparse and discontinuous geological data collected in the field.The method can realize the encryption of discrete point sets randomly distributed in two-dimensional or three-dimensional space.Visualization is the process of mapping the interpolated rule data into a two-dimensional image displayed on the screen by computer graphics technology.Interpolation and visualization technologies have developed rapidly and are widely used in various fields such as geological exploration,meteorology,hydrology,and medical imaging.In the visualization process of near-surface attribute modeling,the following research works were mainly done in this paper:(1)A variety of interpolation algorithms are studied for the proposed near-surface modeling visualization method based on attribute data volume interpolation.Firstly,the anti-distance weighting and the hierarchical B-spline interpolation were studied and implemented.The related theory of spatial statistics was studied and studied,and the ordinary kriging interpolation algorithm was realized.(2)Upscaling multiple interpolation algorithms,ie,inverse distance weighting,hierarchical B-spline,and kriging methods are extended from three-dimensional elevation data interpolation to four-dimensional attribute data volume interpolation.(3)Researching and realizing a kriging interpolation method based on variable-range sliding neighborhood.Experiments show that this method has obvious speed-up in 3D elevation data interpolation.For the problem that the algorithm can not complete the interpolation of four-dimensional attribute data volume,the paper improves the method.The improved sliding kriging method can successfully complete large-scale attribute data volume interpolation and has good interpolation accuracy.(4)Based on slice rendering and a GPU-based ray-casting volume rendering algorithm,it is applied to the near-surface 3D visualization process,and the near-surface attribute models obtained by interpolation are respectively sliced and volume-draw.The near-surface modelling visualization recreates its internal structure,providing reasonable guidance for deep hole excitation and field static correction for seismic exploration.