Knowledge-Based Control of Vision Systems
Abstract
We propose a framework for the development of vision systems that incorporate, along with the executable programs, the syntactic, semantic and strategic knowledge required to obtain optimal performance. In this approach, the user provides the input data, specifies the vision task to be performed, and then provides feedback in the form of qualitative evaluations of the result(s) obtained. These assessments are interpreted in a knowledge based framework to automatically select algorithms and set parameters until results of the desired quality are obtained. In this manner the vision system is given the capacity to tune itself for optimal performance. A system thus trained on a small subset of the input data can then be run autonomously on the remaining data in a batch mode. This approach is illustrated on two real applications, analysis of Synthetic Aperture Radar (SAR) imagery, and detection of vehicles in aerial photographs.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1997
- Accession Number
- ADA353865
Entities
People
- Chandra Shekhar
- Philippe Burlina
- Rama Chellappa
- Regis Vincent
- Sabine Moisan
Organizations
- University of Maryland