Pyramidal hierarchical relaxation

Probabilistic relaxation has been shown to be a powerful method for extracting features from images [Hancock and Kittler, 1990]. In the present paper, the authors describe an hierarchical approach to the probabilistic relaxation. There are two main ideas in the work. Firstly, the authors use pyramid...

Full description

Saved in:
Bibliographic Details
Main Authors: Wing Hung Lau, Hancock, E.
Format: Conference Proceeding
Language:English
Subjects:
Computer science
Dictionaries
Feature extraction
Filters
Information filtering
Information processing
Labeling
Refining
Relaxation methods
Shape
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Probabilistic relaxation has been shown to be a powerful method for extracting features from images [Hancock and Kittler, 1990]. In the present paper, the authors describe an hierarchical approach to the probabilistic relaxation. There are two main ideas in the work. Firstly, the authors use pyramidal constraints for the extraction of major line features. Secondly, they partition the dictionary items according to the angle formed by the labels in each of the dictionary items to reduce the processing time for traversing the dictionary. The new method is proved to be more efficient than the original method and it also produces a more refined ridge map.< >
DOI:10.1109/SIPNN.1994.344798