A method of parallel-hierarchical transformation based on population coding is described and the use of this method for image recognition is considered. This parallel-hierarchical transformation is described as a system model for image recognition. Theoretical information, experimental investigations, and a software realization are presented.

No abstract

No abstract

Paper deals with organization principles of parallel-hierarchical (PH) transform for multistage perception and processing, compression and recognition of information in computational systems which make use of computational scheme similar to neural. Unified methodological approach was developed for analysis of parallel processes. This approach considers influence of structural hierarchy in dynamics. It tracks processes of spatial areas transformation of correlated and generation of uncorrelated in time elements of generated PH network, at the time of transition of the network from one stable stage to another. The main feature of the proposed method is the studying of the dynamics of spatial-correlated procedure for transformation between current and output units in PH network. This procedure gives an opportunity to present processing in PH network as the process of parallel-sequential transformation of diverse image components and consideration of temporal characteristics of transformation. The features analysis of structural-functional organization of networking model of PH transformation of the information environments and corresponding masks generation methods is realized in the given research. On the base of research carried out the method of optimized masks formation while information coding in PH transformation is proposed. The software package, characterized by increased efficiency of direct and reverse PH transformation is developed

No abstract