Handwritten English Letters Recognition Based On BP Neural Network

With the development of society,English as an international common language has been increasingly widely used,so a large number of documents in English need to be tidied up,queried and statistic,and the English Document Recognition system can easily complete many unimaginable work before.Handwriting Recognition System for English letters can be divided into three main parts: input image,normalization and recognition. The English letter samples used in the paper have been got from the paint board belongs to Windows system.This paper uses the imread.m function of MATLAB to read the handwritten letter images, because the letter images have not yet reached the feature extraction request, we need use the self-defined normalization method to preprocess 28×28 pixel images to10×14 pixel images,then we extract features pixel by pixel.After pre-processing we encode the gained letters characteristic which used to be input of the neural network. By cramping out the characteristic vector from samples, we select 520 samples(20 groups) to train the BP Neural Network, and then put the other 520 samples into the trained Neural Network, the distinguished digital scripts are obtained according to the output of the neural network, and we carried on simulation in the environment of MATLAB.This paper shows that Handwritten English Letters recognition system based on BP neural network is able to effectively recognize characters in the same font of training samples. In addition, it has certain capacity of anti-interference and anti-deformation, it can also be used in other fields related with OCR.The designing principle, process and the program of BP net and some source program will be shown in this paper.