Universal binary and multi-valued neurons paradigm: Conception, learning, applications

Futheron development of the Multi-Valued and Universal Binary Neurons conception with basic arithmetic over Complex Numbers Field is presented in this paper. Lot of attention is devoted to Universal Binary Neurons. New high-effective fast convergenced learning algorithm based on Error-correction rule is considered. It is shown that any non-threshold Boolean function can be implemented on the single Universal Binary Neuron. Example for solution of the XOR-problem on the single neuron is considered. Applications of the UBN for solution of the important problems of Image Processing (impulsive noise detection and filtering and edge detection with extraction of the smallest details based on representation of these operations through non-threshold Boolean functions) are also considered.