Scene Analysis Using a Semantic Base for Region Growing
STANFORD UNIV CA DEPT OF COMPUTER SCIENCE
Pagination or Media Count:
The problem of breaking an image into meaningful regions is considered. A probabilistic semantic basis is effectively integrated with the segmentation process, providing various decision criteria. Learning facilities are provided for interactively generating the Bayesian probabilistic basis. A programming system which is based on these ideas and its successful application to two problem domains are described.
- Computer Programming and Software