Highly Perceptual And Space Adaptive 2D Vector Field Visualization

Flow visualization is an important and classic branch of scientific visualization, which has been the subject of active research for many years. There are many examples of vector field, such as stream flow and airflow. It is hard to observe their inner structure, because vector data is not visible. However, People would like to explore what the abstract vector data looks like indeed. They did analyze, study and summarize vector field features, and discover a technique named scientific visualization which could make the abstract data to be a picture which can be seen, Flow visualization technique is very useful to obtain important information and reconstruct inner structure. A good streamline visualization system should be able to accurately extract the topology direction and speed expression of the vector field.This paper presents an advanced space-based 2D streamline placement algorithm with topology-aware characteristics, loop detection strategy, texture-based enhancement and animation technique. It divides into two parts:Streamline organization and Streamline animation.In the streamline organization part, some key techniques are introduced.(1) Marine grid data based feature extraction technique. Before streamlines begin to integrate, it needs to conduct a global scanning. In this process, the singularities are extracted to keep the primary flow features, especially the vortices.(2) Space adaptive streamline visualization technique. First of all, fourth order Runge-Kutta with Inverse Distance Weight method are applied to make streamline smoother and more accurate. With Cartesian grid’s help, Adaptive distance control is explored to maintain more details in high speed area and less detail in low speed area, which can balance the information expression with time consumption.(3) Loop detection technique. Without effective loop detection, the flow field might be cluttered by closed or spiraling(not closed) streamlines. Even more, it might be a time-costing process. It is more susceptible for the algorithm based on the inter-sample distance control strategy to this problem. Because each sample is checked against only the samples of other streamlines instead of checked against samples of the same streamline simultaneously. In this paper, we apply a dot-product-based method to realize a robust and fast loop-detection function. It could kill the ill-loop in the bud, as well as, keep the balance between time saving and visual optimization.(4) Texture-mapping technique. Direction and magnitude is the most essential two attributes of flow field. General geometric streamline is usually hard to express these two attributes, While, Texture mapping is a good way to enhance streamline’s magnitude and direction. It makes the streamline visualization more adapt to human vision.In the streamline animation part, according to the different situation, such as steady flow or unsteady flow, there are different strategies.(1) Color table technique based geometric flow animation strategy. This is a method special for steady flow. Information doesn’t change in steady flow as time goes by. It needs to periodically update color table of streamline to realize animation.(2) Correlation technique based geometric flow animation strategy. This is a method special for unsteady flow. In the circumstances, flow information vary from time to time. In order to keep the fluency when the animation is under way, It is actual to keep the streamline correlation, as well as, to keep the texture correlation between the reference and corresponding streamline.All the algorithms in this paper can be achieved for large flow fields on an ordinary PC. Our algorithm conveys more features and is more perceptual but not more time-consuming than other approaches through analysis with compare.