| OCR Text |
Show 2 data; earth-bound radio astronomy arrays; instrumental arrays recording geophysical entities, such as ocean temperatures, ocean floor features, tectonic plate and volcanic movements, and seismic reflections from geological strata; and medical scanners employing various imaging modalities, such as computed transmission and emission tomography and magnetic resonance imagery. Two fundamental algorithms used to visualize scalar field datasets are direct volume rendering algorithms and surface fitting algorithms. The direct volume rendering algorithms include approaches such as ray-casting and splatting methods. These methods are described in detail in Chapters 2 and 3. One of the disadvantages of direct volume rendering is that the entire dataset must be traversed each time an image is rendered. The amount of data that needs to be processed first in order to find the relevant visual information and to interactively explore it is very large. Often, scientists are interested only in a part of large and complex dataset. Complexity of data and data gathering modalities result in structures obscuring the regions of interest. All these factors pose an interesting problem to the computer scientist. In some cases, data signal is distorted by noise or the scientist is interested only in part of the signal. It is important that the scientist be able to explore the data quickly. The scientist should be able to view the primary visualization results in a reasonable amount of time. Direct volume rendering and surface fitting algorithms as stand alone systems cannot meet the above mentioned needs of a scientist. The thesis Matte Volumes for Selective Data Visualization aims to meet some of these needs. A Matte volume is a function defined on a volume specifying a scalar value at each point in the volume. This value can be interpreted as the percentage of voxel contained by the matte or percentage of the material of interest contained by the matte. Matte volumes can be stored as an explicit array of values or by a procedural definition. Matte volumes allow scientists to process and manipulate 3D scalar field data to produce better scientific visualization. The system provides the scientist with tools for manipulating the data at reasonable speeds and for focusing the attention on a particular portion of the data. It also attacks the problem of rendering of |
