Geo*View

February 1, 2006

Visualization of geospatially correct, remotely sensed data is a key element of many government and commercial applications. It enables a user to analyze and assess ground activities and other conditions of interest. Because remotely sensed data can include a diversity of data types reflecting many different data formats, users may experience difficulty visualizing and interpreting these varying data types and formats due to data structure complexity. In addition, important supplemental information often accompanies the data. This supplemental information—or metadata— may include pertinent information of significant value to the user with respect to where, when, and how data collection occurred. Whereas some applications require metadata to support geospatial analysis functions such as positioning and measurement, many others are unable to interpret such metadata and it may thus go unnoticed. Multiband data and motion imagery further compound the task of visualization with spectral components and complex video streams interlaced with other geospatial information.

Figure 1. Typical Geo*View terrain visualization

AFRL, in collaboration with PAR Government Systems Corporation, developed Geo*View, a general-purpose, easy-to-use, point-and-click-oriented imagery viewer that supports a variety of visualization requirements for government and commercial remote sensing applications (see Figure 1). Geo*View is a Java™-based, platform-independent, self-installing application with a plug-in architecture designed for timely integration of remote sensing, geospatial, and image exploitation algorithms. The viewer supports commercial imagery, multi-hyperspectral imagery, Moving Picture Experts Group (MPEG)-2 video, standard mapping and geospatial products, and Department of Defense (DoD) imagery formats. Hyperspectral aids provide automated band selection, plug-ins for anomaly detection, and spectral-matched filtering. Other developmental features include a local spectral signature library, band reduction and aggregation functions, and pixel classification.

Motion imagery support provides basic controls for video play, pause, and frame advance. Frame-based MPEG-2 video multiplexing/demultiplexing and metadata decoding are available for parsing out geolocation information. The viewer also features basic operations for tracking and annotating objects, building video mosaics, and clipping out frames. Geo*View is National Imagery Transmission Format (NITF) 2.1-certified to complexity level 7 for reading and writing. It autopopulates and validates the NITF geospatial metadata and supplies an NITF metadata editor for updating and regenerating NITF-compliant data files (see Figure 2).

Figure 3. Mapping toolkit 3-D representation

Reusable software toolkit libraries for mapping and three-dimensional (3-D) visualization provide capabilities that can interact with Geo*View or be integrated into other viewing applications. Operators can configure the mapping toolkit to act as a map-based query function or a viewable canvas for layering maps, imagery, graphics, and shape files in a common geospatial reference. They can also use the 3-D visualization toolkit to load, display, and drape map data, imagery, and video over digital terrain elevation data to create a 3-D representation of a region of interest, which they can then manipulate to change the viewing perspective (see Figure 3). AFRL has distributed Geo*View to numerous government agencies and users and affiliated developers for on-site use and integration with other DoD systems and capabilities.

Mr. Frederick W. Rahrig, of the Air Force Research Laboratory's Information Directorate, wrote this article. For more information, contact TECH CONNECT at (800) 203-6451 or place a request at http://www.afrl.af.mil/techconn/index.htm . Reference document IF-H-05-10.