Method for training of a parallel-hierarchical network, based on population coding for processing of extended laser path images

A method of organization of a parallel-hierarchical network for image recognition

“…In Fig. 7, curve 1 (method 1), curve 2 (method 2), and curve 3 (method 3) represent, respectively, coding based on the Q-decomposition [13], population coding [8], and population coding with comparison of masks [6,13]. Thus, the results of correlation comparison for PI network levels for different face images coded by method 3 (curve 3) do not correlate among themselves (the average correlation coefficient does not exceed 0.2).…”

“…It is obvious that, at the level of branches of a neural network, this finite action corresponds to averaged parameters of this network if population coding is realized [8]. In this case, the current image being recognized is represented by the current PI network and is compared with a reference PI network with averaged parameters a NM a…”

“…(6) perform tunneling over averaged correlation coefficients in comparing the reference pattern with a separate trace fragment (unity is taken as the upper bound of the tunnel, and the averaged correlation coefficient is taken as the lower bound); (7) determine "good" or "bad" patterns (a trace fragment is considered to be good if the average correlation coefficient belong to the tunnel of the reference pattern and, otherwise, the fragment is considered to be bad); (8) optimize weights for a good pattern using the formula …”

Application of parallel-hierarchical transformations for rapid recognition of dynamic images based on GPU technology

“…This problem can be solved through application of the laser-based technologies [1][2][3]. One of the most efficient methods of large data arrays processing is their parallel processing based on the neurallike parallel hierarchical systems.…”

Modeling of a method of parallel hierarchical transformation for fast recognition of dynamic images

“…Large volumes of information may be transferred easily through a fiberoptic cable, but laying such cable is rather expensive, even on short distances. At the same time, this problem can be solve using laser-based technologies [1][2][3], one of the most promising models of information transfer for the near future. In this case, for instance, tons of full-length films and virtual worlds could be transferred to any part of the globe in a blink of an eye.…”