一种改进BP神经网络在模式识别中的应用  

Pattern Recognition Using an Improved Back-propagation ANN

作　　者：谢蓄芬[1] 刘泊[1] 王德军 

机构地区：[1]哈尔滨理工大学测控技术与通信工程学院,黑龙江哈尔滨150040 [2]哈尔滨微克科技股份有限公司,黑龙江哈尔滨150090

出　　处：《哈尔滨理工大学学报》2004年第5期63-65,共3页Journal of Harbin University of Science and Technology

摘　　要：为了提高神经网络识别的速度及实用性,在传统BP神经网络的基础上,提出了一种动态调整网络结构的改进网络模型,给出了该模型在应用中的算法,并做了特征提取.仿真结果表明,与传统的BP神经网络比较,改进网络的初始权值的选取相对容易,有效降低了陷入局部极小的可能,提高了网络收敛和识别的速度.For improving the speed and practicability of ANN in recognition, in this paper, a back - propagation neural network model has been proposed to produce a more effective recognition based on adjusting the connection mode of the network. In addition, an algorithm and feature extraction corresponding to the improved ANN also have been proposed. Experiment results show that the weights initialization becomes easier, and he convergence and recognition speed are increased.

关 键 词：模式识别 BP神经网络 局部极小 仿真结果 特征提取 初始权值 算法 速度 实用性 传统 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP183[自动化与计算机技术—计算机科学与技术]