UB Virtual Reality Laboratory

Natural or Man-made Post-disaster Visualization

Goal
The goal of this project is to monitor natural or man-made post-disaster situation with GIS/GPS, having multiple choice of optimal path to nearest hospitals, implementation of large amount of data from information fusion process.

Abstract
The purpose of this project is to simulate disaster and its relief situation for discision support/making in a control center. When disaster occurs, there normally is huge amount of data flowing into the control center and thus data fusion process is needed for efficient relief action, such as data visualization, situation awareness and decision making. Main technique includes geo-referencing GIS data, real-time optimal path simulation and multiple optimal path direction in immersive virtual environment. This multidisciplinary project simulates the post-disaster situation of Northridge, CA Earthquake in 1994.

Screen Shots

Seamless Transition: Casualty clusters are visualized over the map (middle). Then it shows casualty clusters of different severity to visualize the overlapped region in purple, which needs more attention.

Morphing Simulation of Casualty Cluster:
Shows time-aggregated data in post disaster situation (left). Right figure shows a cluster boundary and its cells to represent the distribution of casualties in the boundary.

Visualization Tasks for Information Fusion
The information fusion can be largely categorized into three levels - Positioning, identification(Level 1), Situation Awareness (Level 2) and Threat Assessment (Level 3). The role of visualization task is to provide federates effective displays and user interfaces for efficient decision support/making.

Visualization tasks for fusion levels (left), data flow for visualization (middle), and
data layers for display.

Fast Vector Map and GUI
The fundamental task for fusion visualization is to have geo-referencing systems for positioning and identification of objects. The Fast Vector Map implements visualization based on data layer structure. Unlike other map based applications, such as ArcVIEW or GRASS, its rendering speed is near-realtime, which is suitable for disaster monitoring.

Application of data layers for Fast UTM Vector Map of CA State: Zipcode areas (left),
with random colors (middle), and major streets.

GUI: Simulation time controls (left), snapshot of the visualization (middle), and
report monitoring console. Programmed with C# .NET in Windows XP.

Dynamic Symbology
Symbols play an important role in identifying informative objects and are widely used in geo-spatial decision support systems and applications. In high level fusion applications, however, simply placing symbols often lead to information overload problem; symbols quickly grow fast in many applications, such as the post disaster monitoring system we are interested in. This leads to cluttered and overlapped icons. With today's advanced technologies, new visual effects can lead to better visualization systems where iconic overload may be perceived as a problem. Therefore, conventional method of storage-indexing-retrieval of large sets of prepared icon images is not flexible enough for the visualization of higher fusion levels. Instead, we propose a dynamic symbology, which automatically generates symbols from parameterized components in a three-dimensional space. The extension to tactical graphics can provide better situation awareness from simplified and abstract visualization.

Symbol parameters from fusion process (left), software architecture (middle), and
icons of 100 uncertainty levels generated in run-time with one iconic image file.

Seamless Display
The seamless display of higher fusion level can help in situaltion awareness. However, most applications, such as digital tactical maps or battle field visualizations, usually have no relationship between tactical symbols and tactical graphics. Hence the design of dynamic symbology involves interaction between two symbol categories to obtain smooth transition. When connected to fusion outputs, it will display seamless transitions that give a user both detail identity and aggregated information. For example, cluttered casualty symbols in our hazard fusion map can be grouped and represented as areas of severity levels.

People
Dr. Peter Scott (Dept. of Computer Science Engineering)
Dr. T. Kesavadas (Dept. of Mechanical Engineering)
Youngseok Kim (Dept. of Mechanical Engineering)
Matthew Mandiak (Dept. of Mechanical Engineering)
Pritul Shah (Dept. of Mechanical Engineering)
Venkataraghavan Gourishankar (Dept. of Mechanical Engineering)

Affiliates
Center for Multisource Information Fusion(CMIF)

Sponsors
US Air Force, Post-disaster relief project through a $2.5 Million Five year grant.

Publications
[1] Kim, Y. S., and Kesavadas, T., "Automated Dynamic Symbology for Visualization of High Level Fusion", Fusion 2004, Stockholm, June 27- July 1, 2004.