A 2-PHASE CONNECTIONIST APPROACH TO INVARIANT PICTURE INTERPRETATION

The paper presents an efficient two-phase approach to picture interpre tation based on original connectionist techniques. During the first ph ase invariant representations of individual objects are obtained based on third-order image correlations and appropriate neural network clas sifiers are used to provide a probabilistic assignment of labels to ob jects. The second phase uses relationships between objects to reduce o r eliminate ambiguity by means of a relaxation scheme based on stochas tic 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.