基于神经网络的字符识别

haracter recognition in the field of pattern recognition is a traditional issue because of the character recognition is not an isolated problem, but in the field of pattern recognition in most of the encounter will be the subject of fundamental issues. Human-character recognition to speed up the exchange of information is an effective means. At present, there are a lot of information in the form of books, by hand if the entry way, not only the efficiency of the ground floor, and error-prone. In such a request, with character recognition of the need arise. In this article mainly to the use of neural networks to achieve control of pollution in a 5 × 5 image of the recovery, identification, in order to achieve character recognition. The main use of recurrent neural network (recurrent network) in the Hopfield network to achieve.