Rapid Terrain Visualization; Meeting the Need for Contingency Data Sets

The Rapid Terrain Visualization program will demonstrate rapid production of high-resolution digital topographic data using airborne Interferometric Synthetic Aperture Radar (IFSAR) and Light Detection and Ranging (LIDAR) sensors on a de Havilland-7 aircraft. In a parallel effort using ground-based workstations, feature data is generated from multispectral imagery, IFSAR, and LIDAR data, with conflation of overlapping coverages. U.S. Army topographic specialists and joint users will extract and export feature data in ARC Coverages, standard NIMA formats, or simulation terrain data formats. RTV software will greatly accelerate production of terrain data in the field to rapidly respond to contingencies and natural disasters.

Background. In the Post-Cold War Era, the mission of U.S. ground forces is transitioning from one based on forward deployment to one based on force projection. A crisis or famine in some distant land may warrant deployment on very short notice, and often there is limited terrain data available over that area of operations. The Rapid Terrain Visualization (RTV) Advanced Concept Technology Demonstration (ACTD) was initiated in 1996 to demonstrate technologies that meet Army requirements for rapid generation of digital terrain data. The RTV ACTD is managed by the Joint Precision Strike Demonstration Project Office, Fort Belvoir, VA.

Data Generation Requirement. The RTV ACTD is striving to meet the U.S. Army's data generation requirements shown in Figure 1, which includes high-resolution elevation data, imagery, and feature data. The RTV goal is to demonstrate technologies that can be used to supplement and complement production by the National Imagery and Mapping Agency (NIMA).

• 20km X 20km area in 18 hours
• 90km X 90km area in 72 hours
• 300km X 300km area in 12 days

RTV Concept. RTV processes for rapid generation of terrain information are shown in Figure 2. RTV processes include:

• Collection and processing of high-resolution elevation data and imagery
• Generation of feature data from various kinds of source data

Collection and Processing of High-Resolution Elevation Data. RTV employs two state-of-the-art sensors to acquire high-resolution elevation data: Interferometric Synthetic Aperture Radar (IFSAR), and Light Detection and Ranging (LIDAR). Both sensors will be operated from RTV's de Havilland (DHC-7) aircraft.

• The LIDAR sensor is capable of generating Level 5 elevation data (1m post spacing). Additional LIDAR products include a color-coded elevation image and a black and white intensity image. This highly accurate product supports detailed planning and visualization for missions such as hostage rescues. LIDAR data can also be used for detailed feature extraction of urban areas because of its resolution, and because heights of features like buildings can be measured directly.

• The all-weather IFSAR sensor will be used to collect Levels 3 and 4 elevation data (10m and 3m post spacing). Additional IFSAR products include an orthorectified magnitude image, a correlation file, and a quality file. The RTV IFSAR system collects and processes elevation data in real time on board the aircraft, a significant technological breakthrough. High-resolution elevation data from IFSAR provides a foundation for terrain understanding and topographic analysis, as well as a valuable source for feature extraction.

Figure 4. IFSAR Products

All RTV IFSAR and LIDAR data is provided in GeoTIFF format with a UTM projection, WGS84 horizontal datum, and Mean Sea Level (orthometric) vertical datum. WGS84 vertical datum can also be provided if requested.

RTV Data Generation Software. This software package consists of four modules: Terrain Data Manager, Multispectral Imagery Feature Extraction, IFSAR Feature Extraction, and Feature Attribution.

• Terrain Data Manager (TDM) provides for easy ingest, display and query of available geospatial data. The user-friendly interface supports data query by type or area, web access to search and download data from other TDM sites, and metadata management. TDM launches the other three software modules and tracks the status of the data generation process. The TDM module is built around Oracle software for database management and Orbix software for communications management. TDM also includes a tool to export elevation or feature data in SEDRIS Transmittal Format to support simulation terrain databases.
• MSI Feature Extraction (MSI FE) performs semi-automatic or manual extraction from several types of multispectral imagery (Landsat, SPOT, IKONOS, and ADAR-5500). The software requires source data to be rectified, and to be formatted in ERDAS Imagine format. Extracted features are assigned codes from the Feature Attribute Coding Catalog (FACC). The operator has a full suite of easy-to-use editing tools, and can display other source data such as imagery or raster maps to assist in the extraction process. When the operator is satisfied with the vegetation, soils, and lines of communication layers, the data is sent back to TDM as ArcGen vector files.
• IFSAR Feature Extraction (IFSAR FE) performs semi-automatic extraction from RTV IFSAR data similar to the MSI FE module described above.
• Feature Attribution (FA) enables users to augment the feature and attribute content of data generated in the two FE modules. FA is built on ERSI ARC/INFO and ARC GRID, and includes all of the sophisticated capability of those commercial software packages. FA's strength comes from its ease of use and process-oriented workflow design. It allows users to build vector data sets to NIMA specifications with pop-up menus and pre-defined attribute value fields. New data sets can be constructed from scratch using the ArcGen files brought in from the FE modules, or the FA module can be used alone to update existing NIMA data sets. When the operator is satisfied with the feature accuracy and level of attribution in each of the layers, data can be exported in a variety of NIMA formats to include Vector Interim Terrain Data (VITD), Digital Topographic data (DTOP), and Foundation Feature Data (FFD). Data can also be exported as Arc Coverages or Esri Shapefiles if required. The FA module also incorporates tools that allow the operator to extract features from RTV LIDAR data and export them as a NIMA Urban Vector Map (UVMap) data set. The FA module includes a state-of-the-art capability for automatic conflation of feature data from multiple data sets. The conflation function is derived from the Multiple Database Integration and Update software developed by Northrup-Grumman, Inc., which utilizes ERSI Spatial Database Engine (SDE) and Oracle software.

Figure 5. RTV VITD Results From High Resolution MSI

Several iterations of the RTV Data Generation Software have been built and improved, with a typical turn-around of six to nine months. Each prototype software package was tested by soldiers to make sure it met the needs of Army Topographic units and was delivered to topographic units at Fort Bragg, NC and Fort Hood, TX.

Schedule. The RTV ACTD is presently in its fourth year of execution. All development on the RTV aircraft and RTV software will be completed by October 2000. Throughout FY2001, the RTV ACTD will be working on several areas:

• Operational data collection of IFSAR and LIDAR
• Transition of the DHC-7 aircraft and sensor technologies to the user community
• Transition of the Data Generation software to the Project Director, Combat Terrain Information Systems (CTIS), for integration into the Army's Digital Topographic Support System – Base (DTSS-B).

Conclusion. The RTV ACTD has integrated and demonstrated tools for rapid generation of high-resolution elevation and feature data. RTV capabilities will enable ground forces to rapidly generate digital terrain data when unexpected contingencies arise over areas of the world which lack sufficient map coverage.