Research And Implementation Of Vector Field Visualization

Vector field visualization technique is one of the most challenging application domains in scientific visualization. One of the main reasons is that no natural visual representation exists for vector fields, especially for three-dimensional vector fields. Vector field visualization provides a powerful means for researcher to understand the complex phenomena hidden in the gigabytes of data. This dissertation puts its emphasis on the research and implementation of vector and tensor field visualization. Focusing on the visualization techniques based on lines and surfaces, local velocity visualization based on tensor decomposition, and feature-based time-dependent data field visualization, the dissertation consists of the following parts.1. For structured grid data field, the dissertation gives a point location approach for the generation of stream lines and stream surfaces. This approach finds the nearest grid point of the given point and determines which grid cell contains the given point. The discrete points on stream line and path line are directly generated in physical space . For unstructured grid data type based on tetrahedral grid cells, the point location operation is performed directly in physical grid cells. These methods have been used in stream surface construction. When stream surface diverges greatly, a particle is added in order to smooth the surface. The position of the added particle is determined by the point location approach. A surface tiling method is also discussed for smoothing the generated surfaces.2. The dissertation also presents a visualization method called the deformed cube for visualizing three dimensional velocity vector field. Based on the decomposition of the tensor which describes the changes of the velocity, it...