Research On Meteorological Scalar Data Visualization Technology

Meteorological information is a kind of key geographic information data,which consists of scalar and vector data.In general,most of the meteorological data obtained through measurement or simulation appears in the form of scalars,so the visualization of meteorological scalar data is of great significance to meteorological analysis and application.This thesis introduces the research status and methods of scalar visualization in detail,takes meteorological scalar data as the main body,outlines the processing methods and visualization methods of meteorological scalar data,in addition,the scalar data interpolation method and the contour tracking and labeling method are carried out with three-dimensional data visualization method.In-depth research and analysis.The main work and results are as follows:(1)A neural network interpolation model based on the optimization of beetle whiskers is proposed.The model uses national meteorological scalar data as experimental data,and conducts comparative experiments with neural network models such as genetic neural network,particle swarm neural network,and BP neural network.The experimental results show that the neural network interpolation algorithm based on the beetle whisker is effective and fast.Secondly,the experimental data results under the interpolation model are compared with the data of the traditional meteorological interpolation algorithm,which further proves that the neural network interpolation model based on the beetle whisker is effective and accurate in meteorological scalar interpolation.(2)In the process of isoline visualization,isoline tracking is crucial.When the data size is large,the traditional tracking method is time-consuming.To solve this problem,this thesis proposes a method based on four-sided contour tracking.Through comparative experiments on different types and different magnitudes of data,it is proved that the four-sided tracking method is much faster than the traditional tracking method.Then,a method for determining the contour labeling position based on slope and step size is proposed.The results show that the labeling effect is more obvious.(3)A data classification method based on clustering is proposed,which establishes a three-dimensional visualization model by combining color and data height.Solved the problem of poor visualization of 3D models.(4)Based on the previous work achievements,a visualization platform for meteorological scalar field was designed and implemented,in which meteorological scalar data processing and visualization effect display were completed.