Visualization Techniques for Supporting Rapid Decision-Making Involving Large Scale Data

Collaborating with scientists and first-responders, Jason Leigh, Luc Renambot, Andrew Johnson, and Thomas DeFanti have identified a real and emerging need for a new generation of tools that will facilitate the rapid and collaborative visual fusion of extremely large, real-time data sets. Specifically, there is a need to visualize extremely high-resolution, two-dimensional imagery data such as those from aerial photography for homeland security applications. Also of concern is visualizing large three-dimensional volumetric data such as those collected from seismic probes in deep sea core drilling expeditions.

The team members are developing JuxtaView and Vol-a-Tile, tools that will support large data visualization on scalable resolution displays. JuxtaView will allow users to fuse together data and visualizations from a variety of data sources. The software will also allow them to collaboratively interact and annotate enormous image databases stored on remote data servers that are connected gigabit networks. The team, located at the Electronic Visualization Laboratory (EVL) at the University of Illinois at Chicago, will also house a collection of image datasets at the StarLight facility in downtown Chicago and stream the data on-demand to TRECC via I-Wire. Experimental datasets will include US Geological Survey (USGS) homeland defense, aerial photography of the entire Chicago area, and core drilling samples provided by the Joint Oceanographic Institutions. JustaView has immediate applications in planning evacuation routes for major cities and in oceanographic and medical research.

Vol-aTile will complement JuxtaView's two-dimensional visualization capabilities as a collaborative visualization of enormous time-varying three-dimensional volume data sets on scalable resolution displays. Vol-a-Tile will allow multiple users at distributed sites to navigate through arbitrarily-sized data volumes and to apply complex color/opacity transfer functions to reveal intricate structures hidden in the data. Vol-a-Tile will be a useful tool in geoscience and oceanographic research as well as medical imaging.