Application of BP Neural Network in the Digital Recognition System

摘要

the purpose of digital identification technology based on artificial neural networks is to achieve efficient character recognition through the simulation of human brain function and structure. This paper uses BP neural network to design three classifiers which separately extract the input vector with different characteristics,proposes a weighted integration multi-classifier method based on the analysis of common used integrated multi-classifiers,and uses the method to integrate the designed three classifiers to design a weight-based integrated multi-classifier handwritten digital identification system.