Visualization And Feature Extraction Of Ocean Flow Field Data Based On GPU

With the continuous development of information science and technology,massive data is increasingly hitting people's sights.How to use science and technology to present data in an intuitive form is the current research focus.Therefore,visualization technology is used,and the visualization of flow field data is one of the research hotspots,and has broad development prospects and application value.In order to effectively solve the problem of low efficiency of complex flow field visualization and to speed up visualization,this paper proposes a GPU-based ocean flow field visualization algorithm.In addition,vector field data is the representative data in the field of scientific visualization.How to extract the feature information in the data and visually display of the inherent rules of the data is the focus of vector field visualization.To this end,this paper proposes a feature extraction method based on information entropy,extracts the characteristics of ocean flow field data by using the characteristics of information entropy,and designs a feature visualization method for ocean flow field data.The specific work is as follows.Firstly,the osgEarth 3D virtual earth platform is built to load the ocean flow field data.According to this virtual platform,a set of feasible schemes for realizing the visualization of ocean flow field data is presented,and the visual display of the research object is successfully realized.The format of the flow field data selected in this experiment is NetCDF format,and C++ language is used for programming.Two methods are used for visual expression.The first one is based on the arrow method of geometric shader,and geometric shader is introduced in the process of flow field visualization.The GPU's parallel processing capability and powerful image processing capability are used to draw the real-time arrow of the flow field.And linear color mapping is used to complete the visualization of the flow field data,which shows the distribution of the ocean flow field more completely and intuitively.The second method uses the streamline method for visual expression,and utilizes the powerful graphics computing power of the GPU to greatly shorten the calculation time of visualization.In the arrow method,the integral color mapping method is introduced to deal with the problems that the color is single and the distribution is uneven.The referred mapping method,compared with the conventional linear color mapping method,improves the uniform distribution of colors and enhances the layering of the flow field intensity.Research shows that this algorithm has a better visualization effect on the overall distribution of the flow field,and reduces the amount of data transmission during the visualization process,reduces resource waste,and improves visual rendering efficiency.Then,for the problem of feature point omission detection in the process of vector field data visualization,a feature point detection method based on local maximum entropy is proposed.The local maximum entropy value is calculated in the entropy field corresponding to the vector field data and the candidate feature region is determined.After that,it is determined whether there is a zero point in the candidate feature point region according to the linear interpolation principle.Finally,the position of the feature points is calculated by the principle of contours.Thus,the problem of missing feature points is avoided effectively.Finally,the algorithm is combined with the OSG visualization platform to map the data into the 3D Earth through the conversion of the world latitude and longitude coordinates to the screen coordinates,and global macroscopic display is performed on the geographic level.Combining GPU and osgEarth 3D Earth virtual scene,the visualization of ocean flow field data is completely presented.This paper builds a software platform for data visualization,and analyzes and experimentally validates the proposed methods,which proves that the proposed method is feasible.