基于神经网络的线图数据处理模型

A model used in processing of multiline-Drawing Based on Neural net work

提出了一种能够一次处理一组线图数据的新方法,基于神经网络的基本原理,建立了能够处理线图数据的神经网络模型,并且详细的讨论了径向基函数神经网络的具体应用。成为处理线图数据尤其是多线图数据的有利工具。处理过程主要分成三步:(1)采集数据,建立输入、输出矩阵,(2)训练网络模型,(3)检验误差。文中给出了应用实例及误差分析。

We provide a new method which can be used in Processing of Multi li ne-Drawing in this paper,and established a neural net work model by used of the basic theory of neural net work.we discussed the material application of neural net work in detail,and we also discussed its advantage,its defect and in point area.It include three main steps:(1)collect datum,establish the matrix of input and output;(2)train the net work;(3)test the error.Moreover,the application samp le and the error discussion are include in the paper.