Feature Extraction Of Ocean Field Vortices And Visualization Research

The ocean flow field is typical nonlinear dynamical system. The feature structure in ocean not only has an great affect on the generation, evolvement and dynamic mechanism of large or mesoscale oceanographic phenomena, but also has particular meanings in the field of the economy, society and military, and so on. There are vortices in the ocean everywhere, disturbing the stability of ocean. They can make serious damages to marine organisms and fishing, also, it can make great impact to ocean shipping. Accordingly, it is of great significance and scientific value to research the feature of ocean flow field such as vortices.This paper addresses the feature-based visualization theory and does deep research on visualization technique based on streamline. In recent years, with development of feature visualization theory at home and abroad, some research has been applied in large and complex ocean flow data field. But potential ocean feature structure visualization has no significant improvement such as weak vortices detection. Currently, planar vector field decomposition technique is applied in ocean flow field visualization, and the result of weak vortices detection has produced an obvious effects. As a drawback may be seen that after the vector field is decomposed, some vortices may be missed if they are near the boundary of the grid, an algorithm of interpolating mesh boundary incremental construction is proposed. By linear interpolating, compensating triangulation, constructing local grids, the boundary is preprocessor before decomposition, so as to ensure credible extraction of any type of vortices far away from or close to the boundary. The experiment proves that this method has practical value to a certain extent.At present, it is necessary to make very complex calculation to extract feature regions of flow field vortices, with requiring partial derivative or solving function equation. Therefore, it needs to find a rapid and intuitive method to extract the vortices so as to improve the efficiency. On the basis of research and analysis on the existing flow feature visualization, further more, belowing the premise of ensuring its accuracy and visibility, a method of vortices extraction based on Sperner lemma about combinatory topology is put forward in this paper. In this thesis, all grids that belong to core regions are determined by identifying the velocity vector direction of triangle grid vertices. And it focuses on the streamline extraction for flow field by the numerical integration method. The visualization results of the oceanic investigation data can prove the validity and practicability by comparison with the topology-based method for vortices feature extraction. In addition, it also shows the potential of the feature visualization in physical oceanography and hydromechanics. Finally, we use a simple technology to provide multi-scale description of flow field, through the design of scale option factor K, the user can interactively control the interested vortex in certain scale and achieve vortex feature visualization of different scales.