摘要
利用人工神经网络对零件三维表面轮廓进行非接触测量．详细介绍了光学系统、数据获取，如何训练反向传播神经网络（ＢＰＮ）确定系统映射函数，以及零件３－Ｄ曲面测量方法和实例．该方法降低了对测量系统精度的要求。
A non contact measurement method for a workpiece with 3 D surface is introduced by using a kind of artificial neural network Backpropagation Network (BPN). The optical system, data acquisition, and train of BPN are discussed in detail, and the mapping function of the system determined. The method and the example for measurement of 3 D surface are given. The requirement for the system accuracy is reduced, and testing and aligning the whole system in advance accurately are not needed in this method.
出处
《华中理工大学学报》
CSCD
北大核心
1999年第1期78-80,共3页
Journal of Huazhong University of Science and Technology
基金
机械部教育司科技基金
