Center of Excellence for Geospatial Information Science (CEGIS)
Current CEGIS ProjectsAcademic Affiliates
Automated Data Integration
Objective: Determine capabilities and methods for automated integration of The National Map layers transportation, hydrography, elevation, land cover, and orthographic images using existing data sources. Determine limits based on resolution and accuracy for integration and ability to use metadata to achieve automatic integration. Expand current findings and develop a system for integration to establish feasibility.
History: This project originally was funded under the Geography Prospectus and now is in the third year.
Accomplishments: We have developed three different approaches to achieving the objective of this project. First, we used an empirical study to determine the meaning of data integration. We designed a rating scale (1 through 5), plotted various combinations of data that had been overlaid (for example, vector transportation and orthographic images), and had a group of U.S. Geological Survey (USGS) employees rate the level of integration based on position, shape, and temporality. The second approach was to use theoretical cartographic literature and develop the ranges of integration possible based on scale and resolution ratios. The final approach used automated methods to force the vector data to match the raster images through automatic control point location and rubber sheeting and was completed through a contract to the University of Southern California.
Results and Outputs: The empirical study has provided understanding about data when integrated from a visual, qualitative perspective. The theoretical approach has resulted in a working hypothesis using a factor of two for scale or resolution ratios. That is, two datasets with scale or resolution within a factor of two can be integrated. If the ratio is beyond a factor of two, the integration may be possible, but will require significantly more work and probably human interaction. We have several presentations and publications from the project in national and international forums (see http://CEGIS.usgs.gov/data_integration/index.html ). We are preparing another paper to be submitted to the journal Cartography and Geographic Information Science.
Current Status (2007): This original project was completed in fiscal year 2006, but the quantification and extension will require an additional 2 years.
Planned Future Work: The visual/qualitative assessments of integration need to be quantified by distance, direction, and shape measurements. While we have established a base to determine from metadata on scale, resolution, and accuracy whether or not two datasets can be integrated, these initial findings need to be rigorously tested with a variety of datasets. The automated methods for forcing integration of transportation and image data need refinement and need to be expanded to other data combinations.Generalization for The National Map
Objectives: 1) Determine the problems of automated generalization as applied to The National Map; 2) Assess the current state of automated generalization; 3) Develop an overall approach that can be implemented as a solution to achieving the visualization and delivery of an integrated dataset at a specific resolution (scale) from The National Map. This project examines only vector datasets. Raster generalization is handled under the project Multi-Resolution Raster Data for The National Map.
History: Originally funded by Cooperative Topographic mapping (CTM) as a 3-year project. Presently (2007) the project is in the second year.
Accomplishments: We have examined the problems of generalization for The National Map and determined a need to dynamically change scales and content/symbolization as the areal extent changes on the display of the viewer. We compiled a large literature review on generalization, and have developed methods to decrease the level of content of high resolution NHD (1:24,000-scale or larger) to medium resolution National hydrography Dataset (NHD) (1:100,000-scale). Our NHD generalization approach consists of feature pruning and simplification, where feature simplification is accomplished through rule-based feature modifications, and removal of vertices.
Results and Outputs: We have developed software for pruning NHD data based on drainages areas upstream. Upstream drainage area values for NHD drainage network features are preprocess quantities required for pruning, or feature removal. Preprocessing a data layer in preparation for automated generalization is a common practice. We have developed software using Environmental Systems Research Institute (ESRI) tools to automate data preprocessing where necessary. This work has been presented in a couple forums, see ( http://cegis.usgs.gov/generalization/index.html ), and an assessment of the preprocessing method is scheduled for presentation at the American Congress on Surveying and Mapping (ACSM) 2007 conference in St. Louis.
Current Status (2007): The project is in the second year, and has some initial results with good momentum and potential.
Planned Future Work: We need to apply the literature findings directly to The National Map, complete and refine the methods of generalization for NHD, and expand to other vector datasets including transportation, boundaries, and structure outlines (when available).Building an Ontology for The National Map
The goal of the project "Ontology for The National Map" is to specify feature semantics for richer data models. New data models and associated knowledge organization systems for The National Map can translate traditional topographic information into a flexible spatiotemporal knowledge base that can serve many different application areas. Transformation of The National Map database into a comprehensive geographic knowledge base can bring new dimensions to topographic information delivery and revitalize the role of the USGS as the provider of trusted and consistent geographic information and a valuable geospatial integration framework. The USGS Center of Excellence for Geospatial Information Science (CEGIS) conducts research on topographic information, and will play a critical formative role in the development of ontologies, or knowledge bases, for topographic information.