摘要
提出一种基于BP人工神经网络来识别含有噪声的数字字符的方法.通过对BP网络模型进行训练,并选择纯净和带有噪声的数字字符组成样本,借助Matlab7.0实现数字字符识别,识别率为96%,达到了设计要求.
Based on BP artificial neural network, a method for recognition of the number characters containing noise was proposed . Trained with pure samples and noise samples of the number characters, the digital characters were recognied, using with Matlab 7.0 recognition rate reaches 96% ,which meets the design requirement.
出处
《郑州轻工业学院学报（自然科学版）》
CAS
2009年第2期60-62,共3页
Journal of Zhengzhou University of Light Industry:Natural Science
关键词
数字字符识别
人工神经网络
BP算法
number character recognition
artificial neural networks
BP algorithm
