Rule Based Strategies for Image Interpretation,
Abstract
A rule based image interpretation system is presented which is shown to be effective in interpreting complex outdoor scenes. The system utilizes world knowledge to reduce the ambiguities in image measurements obtained from simple interpretation rules. These rules involve sets of partially redundant features each of which defines an area of feature space which represents a vote for an object. The features include color, texture, shape, size, image location, and relative location to other objects. Convergent evidence from multiple interpretation strategies is organized by top-down control mechanisms in the context of a partial interpretation. One such strategy extends a kernel interpretation derived through the selection of object exemplars and regions which represent the most reliable image specific hypotheses of a general object class. The use of exemplar strategies and other top-down strategies results in the extension of partial interpretations from islands of reliability. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1983
- Accession Number
- ADP001207
Entities
People
- A. R. Hanson
- E. M. Riseman
- J. S. Griffith
- T. E. Weymouth
Organizations
- University of Massachusetts Amherst