Computer-Assisted Terrain and Geologic Feature Interpretation from Aerial and Satellite Imagery.

Abstract

In all earlier efforts in constructing prototype expert terrain related systems, knowledge related to the physiographic region of a site and to the spatial pattern of related landforms were not explicitly presented and used. In this research we have identified, named, described, organized and related detailed, 'book-level' knowledge pertaining to physiographic regions (provinces and sections), physiographic features, topographic forms and landforms. Collected, systematized, and defined landform, geomorphologic, topographic, and physiographic indicators. We have developed an object-oriented model for the factual and structural representation of these terrain features. We have also developed a rule-base for representing the strategic knowledge needed for inferring these features from their own indicators. We have provided for the representation of multiple terrain objects at a given interpretive scenario and for bidirectional reasoning for the identification of terrain features depending on the goals of the interpretation at a given time. Eleven scenaria of photointerpretation problem solving were described. They compose three contexts: landform, physiographic and spatial. The presented case studies concern typical terrain of the Basin and Range Province of Southwest USA (Great Basin and Sonoran Desert). The conceptual scheme was formalized and implemented in a knowledge-base resulting in the Terrain Analysis eXpert (TAX-4-5) system which assists step by step the user in the eleven problem solving scenaria.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1996
Accession Number
ADA326081

Entities

People

  • Demetre P. Argialas

Tags

Communities of Interest

  • Space

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Case Studies
  • Civil Engineering
  • Computers
  • Drainage Basins
  • Ecology
  • Engineering
  • Expert Systems
  • Geography
  • Landforms
  • Pattern Recognition
  • Polyethylenes
  • Remote Sensing
  • Ridges
  • Terrain
  • Topography
  • United States

Fields of Study

  • Geology

Readers

  • Artificial Intelligence
  • Wetland-Land-Environmental Management.

Technology Areas

  • AI & ML
  • Space