Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Volumetric visualization of scientific data

Thesis/Dissertation ·
OSTI ID:6779157

Computer graphics techniques for visualization of three-dimensional objects are now becoming available to the general scientist, through tools such as MOVIE.BYU and PLOT3D. But many scientists are confronted with space-filling volumetric data which even the most sophisticated surface rendering and shading techniques cannot fully illustrate. This thesis identifies the basic requirements for volumetric visualization, demonstrates the feasibility of volumetric techniques, even on very modest hardware, and demonstrates when and why volumetric visualization can provide insights into data that are otherwise difficult or impossible to obtain. Space-filling data contains information distributed throughout a volume. Examples include distribution of spaces in partially mixed grasses, velocity of airflow in the vicinity of a wing, and charge density in a semiconductor device. Also, medical imaging produces enormous amounts of volumetric data which map biological tissues. These data do not always lend themselves to interpretation through surface images, since much of their significant information content cannot be expressed as success. Displaying volumetric data requires a visual representation of data as diffuse, partially transparent structures. This is challenging because the images have much greater information content than solid scenes, and because the integration through partially transparent materials introduces depth ambiguity and reduces contrast. Through the use of smooth motion and careful choice of visualization parameters, volumetric visualization can successfully be used to produce informative illustrations of space-filling data, with minimal introduction of artifact. The method exploits our natural ability to understand three-dimensional structure through motion, minimizing the need to interpret abstract or non-geometric representations.

Research Organization:
Princeton Univ., NJ (USA)
OSTI ID:
6779157
Country of Publication:
United States
Language:
English