Texture-based Visualization Method For 2D Flow Field Adaptive To Vortex Features

Flow field visualization is an important means to explain and describe the dynamic evolution of geographical processes such as atmospheric movement and ocean current movement.With the rapid development of earth system models and intensive large-scale computing systems,the flow fields from earth ecosystem simulations and precise climate analysis are getting larger and larger.In the topological structures of flow fields,the vortex feature is the most concerned by researchers.How to display the typical,core and prominent vortex features in the large-scale dense flow fields scientifically,reduce the impact of redundant and low-value information in the flow fields visualization and calculations,and realize the efficient visualization of important features has become the key issue to data integration and comprehensive application of flow fields.At present,there are a large number of flow field visualization methods at home and abroad,including direct visualization methods,geometric visualization methods,and texturebased visualization methods.Among them,the direct visualization method is based on point icons or color coding,which will cause icon aliasing or lack of features,and it is difficult to reveal the inherent continuity of the flow field.The geometric visualization method is based on streamlines,flow surfaces,etc.However,a specific seed point placement mechanism is required to achieve feature fidelity,and the visual confusion caused by the aggregation and coverage of curves or surfaces will cause feature loss.The texture-based visualization method displays the full picture of the flow fields in the form of a texture image,which has the continuity of the image space,can describe in detail and visually express the global direction information of the flow fields.However,there are still many key technical problems in texturebased visualization methods that need to be studied and resolved,including the measurement of flow field information,the problem of feature extraction,the low visualization efficiency,and the poor contrast of the visualization results.In order to solve these problems,this thesis focused on 2D flow fields,studied the issues including information measurement of vortex feature,texture visualization focused on vortex and enhancement of texture.The main research achievements are as follows:(1)To solve the problem of information measurement of vortex feature and vortex description,this thesis analyzed the limitaions of critical point theory and information entropy in the application of vortex description,and proposed rotation distance as the quantitative basis of voricity information.(2)To solve the problem of poor discrimination and low efficiency in texture-based visualization of 2D flow fields,this thesis constructed multi-frequency noise based on rotation distance as the input texture of line integral convolution,which could display the difference of vorticity information by the difference of texture resolution.Meanwhile,this thesis designed dynamic adjustment scheme for integration step based on rotation distance when calculation the integral streamline,which could improve the fineness of vortex regions and reduce the computation of other regions.(3)To solve the problem of low constrast of line integral convolution texture,blurred image result of small vortex and limitation of visual information expression,this thesis enhanced the line integral convolution texture by gabor filter in the vertical direction of streamlines for vortex features of different scales,and combined texture with scalar attribute fields such as rotation distance by line color mapping.(4)Based on the above research results,this thesis designed a prototype system and conducted experimental analysis.A 2D global wind field data was took as an example to conduct experimental analysis to verify the effectiveness of the method.The experimental results showed that the method proposed in this thesis can extract the key vortex features of the 2D flow field,and effectively realize the feature-focused flow field visualization.