摘要
针对传统的手写体数字识别技术的局限性,本文提出了基于小波和RBF神经网络的手写体数字识别方法,即利用小波较强的去噪功能以及RBF神经网络学习快速、容错性较好等优点来解决手写体数字识别的问题。实验表明,该方法的识别正确率较高。
To avoid the limitation of traditional handwritten number recognition methods, a new recognition method of handwritten numbers by using wavelet and RBF neural network is presented in this paper. This paper integrates wavelet＇ s better de - nosing ability and RBF neural network＇ s faster learning attribute to solve this problem. Experiment results show that this new method has better accuracy rate.
出处
《微处理机》
2005年第4期24-25,28,共3页
Microprocessors
关键词
军写体数字
识别
小波
RBF神经网络
Handwritten number
Recognition
Wavelet
RBF neural network
