 |
 |
 |
 |
| |
|||||
|
|
|
Home > News & Events > MITRE Publications > The MITRE Digest > | |||||||||||||||||||
Seeing More with Nonlinear Visualization August 2004
New techniques for looking at images and data promise to help military and emergency response personnel make decisions faster. Combining aerial photos with close-up photos, for example, may help analysts make decisions better by reducing the need to change focus in a high-stress environment. "We hope to demonstrate that these visualization methods can more easily show the patterns and relationships within various types of data," says Dave deMoulpied, Principal Investigator (PI) of a MITRE-sponsored research project called Nonlinear Visualization Techniques. "Nonlinear visualization provides the ability to display both the details and the overview of a scenario simultaneously," says Erika Darling, co-PI of the nonlinear project. "We feel these techniques can result in faster decision-making by reducing the time it takes to navigate from the overview level to the details and back to the overview level," says Darling. Combining overview and details The overview and detail-viewing combination is suitable for a variety of applications. "We're exploring uses for air defense, border patrol, imagery analysis, and intelligence analysis," says Darling. "These methods can be used by high level decision-makers in combat support and supply chain management who need to figure out where all the pieces of their equipment are located," she says. "First responders in homeland defense and emergency management could also use these techniques to maintain an awareness of where their emergency assets are located." An example of a nonlinear visualization technique is a magnifying lens zoomed-in over a point of interest on a large aerial photo. Say you want to locate a specific car in an aerial photograph of a freeway. You can use the lens to inspect the details of the automobiles while maintaining overall context (below).
User Studies The Nonlinear Visualization team is conducting several user studies at MITRE as part of its project. In one study, participants completed tasks using linear and nonlinear (orthogonal and fisheye, below) lenses to view very large images (an aerial photo and a map).
The majority of the users initially had a strong preference for linear magnification lenses rather than orthogonal and fisheye lenses. With additional use, however, many participants began to appreciate the benefits of nonlinear lenses—particularly fisheye lenses. All users considered the ability to offset the lens from the point of interest, in both adjacent and docked modes (below), to be very useful.
In addition to analyzing photographic images, you can use nonlinear techniques to examine other types of data. Another study conducted at MITRE compared the ability to analyze a complex organizational data tree in a traditional tree view (below left) versus analyzing the tree in a fisheye view (below right). A prototype, FocusTree, was developed for experimental purposes to display a hierarchical structure chart. FocusTree is an extension to Piccolo, a Java-based software toolkit for creating zoomable user interfaces that was developed at the University of Maryland.
The study found that participants were more accurate in determining the degree of separation (number of links) between two particular nodes in the fisheye view than in the traditional view. Semantic Lenses In addition to using lenses as magnifiers, they can semantically show other types of information related to the overview. For example, a map or a satellite photo of a town can be connected to a variety of databases. The town's water lines, fire hydrants, and streets can be superimposed in areas of your choice by a semantic lens. This preserves the context while being able to drill-down into multiple areas of interest.
Another semantic lens example shows the spread of disease. "A decision-maker is able to maintain the overview of how SARS spread through the world while simultaneously hovering over any country to get the number of confirmed cases in that country at that given time" says Darling. For air traffic management personnel, semantic lenses could be used to de-clutter their displays. "Their displays are expected to become increasingly cluttered as more capabilities are added," says Darling. "Semantic lenses could show data such as weather, wind, turbulence, and routing conflicts. Rather than cluttering their entire display with that information, they are able to control exactly where they want to see that data." Chris Newbern, a software engineer and co-lead on the research team, notes, "We developed a software prototype built with C++/OpenGL to experiment with various advanced lensing concepts and interaction mechanisms." The team will continue to evolve the experimental prototypes as it explores the concepts behind semantic lensing. —by David Van Cleave Related Information Articles and News |
||||||||||||||||||||
| Page last updated: January 27, 2005 | Top of page |
Solutions That Make a Difference.® |
|
|
