Flow Visualization Based On Rendering Of 3D Surface Texture

Scientific visualization is a subject which has been advanced and developed during the 80s of the 20th century. Nowadays, it has been used widely in many fields, such as medicine, geological exploration, molecular biology, hydrodynamics calculation and finite element analysis. It has been paid more and more attention in these days. The core of scientific visualization is spatial data field visualization. As the fast development of the science and technology, scientific calculation is involved in more and more fields. Its objects are more and more complex. There are also more and more data and calculation volume to deal with. To explore high-performance, acute and easy calculative method for the visualization of large scaled scattered data is a challenge to scientific visualization.Flow visualization is a classical topic of scientific calculation visualization research. It has been widely used in hydrodynamics, weather forecast and explosion data simulation. In flow visualization based on texture, we usually calculate noise texture to present the features of the flow field.Many scientists and experts have been studying flow visualization for long time. During this period, many methods have been proposed, including Line Integral Convolution (LIC) and Spot Noise. These methods have been widely used in many fields. Based on studying the literature of these methods, a new method is proposed in this thesis.Inspired by van Wijk's"Image Based Flow Visualization"(IBFV) and 3D surface texture rendering methods, this thesis proposes a flow visualization algorithm based on rendering of 3D surface texture. By using the new method, Spot Noise can be extended from 2D to 3D, and can combine the advantage of the Lambertian model and Fractal texture models. More visually powerful scenes can be obtained by changing deferent illumination directions.