本文通过对数字特征的分析对传统的BP神经网络识别算法进行改进,采用二值点阵输入网络并且使网络中每个参数调节有各自的学习率并自适应,采用此方法,使特征值提取算法得到优化并且大大减少了迭代次数,试验结果表明,该算法可以快速准确地对网络进行训练并达到较好的识别效果。 In this paper, the traditional BP neural network recognition algorithm is improved through the analysis of digital features. Using binary matrix input network and each parameter of the network is adjusted to have its own learning rate and adaptive. Using this method, the eigenvalue extraction The algorithm is optimized and the number of iterations is greatly reduced. Experimental results show that the algorithm can quickly and accurately train the network and achieve better recognition results.