Novel recognition method of neural network patterns

Abstract

The invention discloses a novel recognition method of neural network patterns, which comprises a learning step and a recognition step. In the steps, a training sample set is set up, and training sample in the training sample sets are divided in to a plurality of blocks; a learning ahead masking SLAM model is set up, and RBF neurons in the model are utilized to obtain a training sample subset of apriority ordered neural network based on the connectivity nature of homologous congeners; the priority ordering for the training sample subset is carried out, and learning neurons are arranged according to the learning sequence; a PORBF network based on the connectivity nature of homologous congeners is set up, and a testing sample is input into the PORBF network; the output of each RBF neuron isdetected, and the activated output neuron with the smallest serial number is found; and the type of the neuron is taken as the decision output of the network. The invention has higher correct recognition rate, and is more suitable for solving the recognition problem of large-scale super-multi-class patterns.

Description

technical field [0001] The invention belongs to the technical field of computers and provides a new neural network pattern recognition method based on the connection nature of similar things of the same origin. Background technique [0002] The research on pattern recognition methods has a history of several decades and many achievements have been made. For example, Fisher proposed to use the known probability density distribution function of two types of samples to design a decision method to separate the two types of samples; Vapnik proposed the concept of "optimal classification hyperplane", and developed the support vector machine on this basis (SVM) method. These theories and methods are based on statistical theory to find decision rules that can divide two types of samples. In these theories, pattern recognition is actually pattern classification. As we all know, with the development of pattern recognition, the correct recognition rate of many pattern classifications...

AI Technical Summary

Problems solved by technology

Large-scale super multi-class pattern recognition problems have high space complexity and time complexity. Traditional Fisher method, linear classification method, piecewise linear classification method, nearest neigh

Images