Expert Systems for Understanding Remotely Sensed Images,
Abstract
This paper discusses some basic issues in the design of expert or knowledge-based systems for understanding remotely sensed images, with particular reference to cartographic feature extraction problems. The many types of knowledge that must be modelled in remote sensing and cartography and the characteristics distinguishing cartographic image processing from other image processing tasks are outlined. The problems of knowledge representation, knowledge-based search, fusion of information from multisensor sources and knowledge-bases, and the resolution of conflicting information are briefly treated. A distributed architecture and a control structure based on a parallel non-directional search algorithm are outlined. Finally, open problems are mentioned. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 31, 1984
- Accession Number
- ADP003020
Entities
People
- B. A. Lambird
- D. Lavine
- L. N. Kanal
Organizations
- University of Maryland