Skip to main content
U.S. flag

An official website of the United States government

S57. Knowledge engineering of topographic features

 

Closing Date: July 30, 2021

This Research Opportunity will be filled depending on the availability of funds. All application materials must be submitted through USAJobs by 11:59 pm, US Eastern Standard Time, on the closing date.

CLOSED

The U.S. Geological Survey (USGS) is the Nation's largest natural resource and civilian mapping agency, delivering reliable and unbiassed scientific information to its customers and affiliates, aided by the National Geospatial Program (NGP) as its largest program in terms of appropriated funding. The NGP’s principal operational organization is the National Geospatial Technical Operations Center (NGTOC), which is responsible for managing The National Map and the geospatial data that populate it. The Mendenhall postdoctoral position will be a member of the NGTOC’s Center of Excellence for Geospatial Information Science (CEGIS). Approximately 5-10 researchers and support staff in the disciplines of cartography, geography, physical science, and computer science form the CEGIS section. The incumbent will focus on artificial intelligence computational methods such as machine learning and, specifically, deep learning, to increase understanding and problem-solving in the field of Geospatial Science, utilizing a high-performance computing infrastructure and geospatial semantic representation. The appointee reports directly to the CEGIS’ Terrain Analysis scientist in the CEGIS Section in the Office of Innovation at NGTOC, and will collaborate with other CEGIS scientists, NGTOC staff, academic affiliates and other government research groups.

The modeling, identification, and extraction mechanisms for topographic features are in part dependent on an understanding of the creation processes (Wilson and Gallant, 2000), and in part on morphometric properties such as shape and size (Pike, 1988). Datasets supporting these mechanisms include digital elevation data from the 3D Elevation Program, the HTMC, and Geographic Names served by The National Map and images from the Department of Agriculture’s National Aerial Imagery Program. The ability to use these data as a source for extraction of terrain features to support spatial reasoning, natural language processing, topographic science modeling, and map generation, depends on a thorough understanding of both the processes which formed the features, and the everyday conceptions of the environment (Smith and Mark, 2003).   

Recent research in the core of artificial intelligence called knowledge engineering suggests a promising future for this work (Arundel et al. 2020). Knowledge engineering attempts to take on challenges and solve problems that would usually require a high level of human expertise or intervention to solve. Paired with semantic technologies, knowledge engineering enables the explicit representation of knowledge and its further processing to deduce new knowledge from implicitly hidden knowledge. Thus, we seek a Mendenhall Fellow to advance the knowledge engineering of topographic features in support of their extraction from multiple data sources for representation as a part of The National Map. The Fellow will be expected to investigate specific research topics such as:

  • Transferring learning of text recognition methodologies, such as that of the NSF Linked Map (Shbita et al. 2020) project, to topographic features.
  • Interfacing machine learning and Semantic Web technologies to create topographic knowledge bases or link to external knowledge bases (for example, LinkedGeoData).
  • Matching profile to plan view of topographic features to leverage available training data. 
  • Managing scale issues in multi-resolution object detection through machine learning.

The incumbent will develop models using machine learning image analysis in conjunction with traditional geographic information systems and image processing software to isolate and extract topographic features from lidar and other data (Li and others, 2012). The procedures for building specific features from process modeling and ontology design patterns will be developed as part of the research. During the past several years, USGS Center of Excellence in Geographic Information Science researchers have developed ontology design patterns for specific terrain and hydrologic features (Varanka, 2011; Usery and Varanka, 2012; Sinha and others, 2014a; 2014b). The extension of this work to model topographic features will allow their effective extraction from various data sources.

Applicants are encouraged to contact the advisor below early in the process to discuss project ideas and how they fit into the overall goals of the National Geospatial Program.

References:

Arundel, S.T., Li, W. and Wang, S. 2020 “GeoNat v1.0: A dataset for natural feature mapping with artificial intelligence and supervised learning. Transactions in GIS. https://doi.org/10.1111/tgis.12633

Li, W., R. Raskin, and M.F. Goodchild, 2012. Semantic similarity measurement based on knowledge mining: an artificial neural net approach, International Journal Of Geographical Information Science, v. 26, n. 8, p. 1415-1435.

Pike, R.J., 1988. The Geometric Signature: Quantifying Landslide-Terrain Types from Digital Elevation Models. Mathematical Geology, 20, pp. 491-511.

Shbita B., Knoblock C.A., Duan W., Chiang YY., Uhl J.H., Leyk S. (2020) Building Linked Spatio-Temporal Data from Vectorized Historical Maps. In: Harth A. et al. (eds) The Semantic Web. ESWC 2020. Lecture Notes in Computer Science, vol 12123. Springer, Cham. https://doi.org/10.1007/978-3-030-49461-2_24

Sinha, G., D. Kolas, D. Mark, B. Romero, E.L. Usery, G. Berg-Cross, A. Padmanabhan, 2014a. Surface networked ontology design patterns for linked topographic data, Semantic Web Journal. http://www.semantic-web-journal.net/content/surface-network-ontology-design-patterns-linked-topographic-data.

Sinha, G., D. Mark, D. Kolas, D. Varanka, B. Romero, C.C. Feng, E.L. Usery, J. Lieberman, and A, Sorokine, 2014b. An ontology design pattern for surface water features, Proceedings, GIScience 2014, Salzburg, Austria.

Smith, B. and D. Mark, 2003. Do mountains exist? Towards an ontology of landforms, Environment and Planning B; Planning and Design, v. 30, p. 4111-427.

Usery, E.L., 2013, Center of Excellence for Geospatial Information Science research plan 2013–18: U.S. Geological Survey Open-File Report 2013–1189, 50 p., https://pubs.usgs.gov/of/2013/1189/pdf/of2013-1189.pdf.

Usery, E.L. and D. Varanka, 2012. Design and development of linked data from The National MapSemantic Web Journalhttp://www.semantic-web-journal.net/content/design-and-development-linked-data-national-map.

Varanka, D., 2011. Ontology patterns for complex topographic feature types, Cartography and Geographic Information Science, v. 38, n. 2, p. 126-136.

Wilson, J. P. and J.C. Gallant, (eds.), 2000. Terrain Analysis: Principles and Applications, John Wiley and Sons, London: 520 p.

Proposed Duty Station: Rolla, Missouri or Lakewood, Colorado

Areas of PhD: Geography, cartography, geographic information sciences, computer science, physical science or related fields (candidates holding a Ph.D. in other disciplines, but with extensive knowledge and skills relevant to the Research Opportunity may be considered).

Qualifications:  Applicants must meet one of the following qualifications:  Research Geographer, Research Cartographer, or Research Physical Scientist.

(This type of research is performed by those who have backgrounds for the occupations stated above.  However, other titles may be applicable depending on the applicant's background, education, and research proposal. The final classification of the position will be made by the Human Resources specialist.)

Human Resources Office Contact:  Audrey Tsujita, 916-278-9395, atsujita@usgs.gov

CLOSED

Was this page helpful?