Visualization Of Spatial Scalar Fields Based On Volume Rendering

In the current era of rapid growth of information data,how to extract and present its intrinsic effective information in massive data is a topic of continuous concern for researchers.Most of the existing data analysis methods remain at the level of two-dimensional analysis,and it is difficult to meet the increasing demand of massive data analysis in quantity and dimension.With the continuous development of scientific data visualization technology,multi-dimensional field data visualization methods emerge in endlessly,and are widely used in various fields.This thesis mainly studies the real-time visualization method of three-dimensional scalar field of meteorological data,using the improved volume rendering algorithm to render.This paper focuses on three aspects of volume rendering kernels,and puts forward the method theory of improving rendering results from efficiency and effect.Aiming at the inefficiency of current volume rendering algorithm,this paper analyzes the four stages of volume rendering: original data processing,data classification,data sampling and data synthesis,extracts the optimal method of each stage,and optimizes the execution efficiency of each stage with the existing mathematical model.(1)Data preprocessing: There are heterogeneous and diverse characteristics of meteorological source data.To unify access to current rendering methods,data standardization is necessary.According to the current data storage characteristics,a custom binary data storage mode is adopted.According to the current data analysis method,the data is parsed and put into the cache space.The data structure is defined,the current data information and data volume are stored,and the current data is serialized into files through binary stream,and the format and storage are unified.(2)Ray-AABB adaptive sampling: In the traditional ray sampling process of volume rendering function,the isometric sampling integration process of ray will cause a large number of invalid sampling.Among them,the calculation of coordinates of ray sampling points is a cross-calculation process of ray and bounding box,which is divided into two parts: out-of-box and in-box.Out-of-box sampling is invalid.So according to the properties of ray and bounding box combined with Ray-AABB detection algorithm of mathematical model,cross-calculation is carried out to find out the current effective ray sampling in the box and incorporate it into the current integration process to realize adaptive sampling.(3)Exponential fog illumination model synthesis effect: aiming at the illumination model synthesis effect in the current synthesis process,its synthesis generally adopts the linear integration process,which is a continuous isometric synthesis,which will lead to the disappearance of the synthesis results when reaching a certain threshold,which is not in line with the current actual human eye observation mode.The key to solve the current critical value problem is to tend to the critical value but not intersect the critical value.In the synthesis process,the exponential function is introduced to simulate the human eye observation mode of the current viewpoint,so as to achieve the real synthesis effect.(4)Shadow synthesis: The current volume rendering synthesis process only uses light absorption and emission effects to synthesize voxels,lacking high-precision light transmission model.In order to optimize the current rendering texture,scattering shadow model is added to the synthesis process.Shadow integration of light source direction is carried out for each sampling point,medium density of sampling point in the box is calculated,and real-time light of sampling point is calculated.Shading.(5)Spatial Scalar Data Visualization System: Currently,meteorological data analysis methods are widely used in forecasting,information and public services of meteorological departments at all levels of the country and provinces,but conventional meteorological analysis methods are mostly in two-dimensional form,which can not simultaneously represent the distribution of multi-layer meteorological elements on the surface,at low altitudes and at high altitudes,and can not reflect the distribution and movement trend of meteorological elements in space.Time dependence.How to analyze and simulate the current geographical environment with three-dimensional spatial scalar data and satisfy the data analysis effect and real-time generation efficiency has become a trend of current meteorological data analysis and visualization.In order to validate the current volume rendering effect,slice data set and three-dimensional scalar data set were used in the experiment,and the phased radar echo of typhoon "Skypigeon" in East Asia was used as data source for visualization efficiency analysis and effect display,and a visualization system based on three-dimensional spatial scalar data was designed.The system realizes the combination of volume rendering classification method and meteorological data classification standard,and cooperates with visual angle control,data slicing and other auxiliary functions to verify the efficiency and practicability of the current algorithm,and provides a good platform for the three-dimensional visualization of meteorological data.