逆向工程中自由曲面自组织重建研究

探讨了曲面密集三维散乱点数据的自由曲面自组织重建方法。建立了基于扩展自组织特征映射神经网络的自由曲面自组织重建模型及其训练算法。所建模型的网络神经元对曲面散乱点的学习来模拟曲面上的点与点之间的内在关系 ,神经元连接权矢量集重构曲面样本点的内在拓扑关系。经过训练 ,神经网络将整个曲面散乱点数据分成许多子区域 ,子区域的分类核心即为神经元连接权矢量 ,每个子区域用一个线性函数逼近 ,实现自由曲面自组织重建。计算机仿真表明 ,所建神经网络模型可实现三维密集散乱点数据自组织压缩及曲面自组织重建于一体。

Research of Freeform Surface Self-Organizing Reconstruction for Reverse Engineering

Based on the improved self-organizing feature map neural network, an approach to the freeform surface self-organizing reconstruction for the dense 3-D scattered data is developed. The inherent topologic relations between the scattered points on the curved surface are reconstructed by the weight vectors of the neurons on the output layer of the neural network. After the neural network is trained, the whole scattered data are divided into sub-regions whose classified core are represented by the weight vectors of the neurons. Every sub-region is approximated by a linear fuction. By this approach, the reconstruction of the surface and the extraction of the large scale data are combined into the same process. The computer simulation results show that this method is effective.