基于BP网络的英文字符识别  

The English Character Recognition based on BP Network

作　　者：吴迪[1] 孙继真[1] 

机构地区：[1]中国人民公安大学安全防范工程系,北京102623

出　　处：《计算机光盘软件与应用》2011年第6期94-94,共1页Computer CD Software and Application

摘　　要：本文利用BP神经网络针对英文字符识别的问题建立模型，通过建立英文字符的图像模板，并给模板赋值形成特征矩阵，最后在MATLAB环境下应用BP神经网络对英文字符进行分类识别，在识别的同时，通过对模板进行模糊处理后再次识别，并与先前识别的结果比较来验证该模型的鲁棒性。仿真表明，该模型能够比较有效地实现英文字符识别，为英文字符识别提供了一定的参考性.In this paper, BP neural network for character recognition problems of the English model building,model images through the establishment of the English character templates,and assign to the template the formation of the characteristic matrix,and finally in the MATLAB environment applications BP neural network to classify the English character recognition,in recognition at the same time,obfuscated by the template again after recognition, and the results compared with the previous identification to verify the robustness of the model.Simulation shows that the model can effectively achieve the English character recognition,character recognition in English provides a reference.

关 键 词：英文字符识别 BP神经网络 MATLAB 

分 类 号：TP391.43[自动化与计算机技术—计算机应用技术]