A two-phase connectionist approach to invariant picture interpretation

 
see the original item page
in the repository's web site and access all digital files if the item*
share




1996 (EN)

A two-phase connectionist approach to invariant picture interpretation (EN)

Kollias, S (EN)
Kontoravdis, D (EN)
Stafylopatis, A (EN)

The paper presents an efficient two-phase approach to picture interpretation based on original connectionist techniques. During the first phase invariant representations of individual objects are obtained based on third-order image correlations and appropriate neural network classifiers are used to provide a probabilistic assignment of labels to objects. The second phase uses relationships between objects to reduce or eliminate ambiguity by means of a relaxation scheme based on stochastic learning automata. Both phases are particularly suited to parallel implementation. Simulation experiments revealed the effectiveness of our approach in solving several problems of small and medium sizes. (EN)

journalArticle (EN)

Correlation theory (EN)
Relaxation Scheme (EN)
Random processes (EN)
Computer Science, Software Engineering (EN)
Simulation Experiment (EN)
Learning systems (EN)
Automata theory (EN)
Invariant picture representation (EN)
Computer vision (EN)
Computer Science, Interdisciplinary Applications (EN)
Learning Automata (EN)
Two phase connectionist techniques (EN)
Computer simulation (EN)
Individual Object (EN)
Neural network classifiers (EN)
Third order image correlations (EN)
Mathematics, Applied (EN)
Object recognition (EN)
Parallel processing systems (EN)
Relaxation processes (EN)
Image analysis (EN)
Neural Network Classifier (EN)
Parallel Implementation (EN)
Neural networks (EN)
Stochastic learning automata (EN)
Probabilistic assignment (EN)


Mathematics and Computers in Simulation (EN)

English

1996 (EN)

0378-4754 (EN)
597 (EN)
613 (EN)
40 (EN)
ISI:A1996UR67100008 (EN)
10.1016/0378-4754(95)00009-7 (EN)
5-6 SPEC. ISS. (EN)

ELSEVIER SCIENCE BV (EN)




*Institutions are responsible for keeping their URLs functional (digital file, item page in repository site)