一种基于特征拟合的网格分割方法及其在反求工程中应用

Feature-fitting based mesh segmentation algorithm and its usage in reverse engineering

随着3D数字化技术的逐步普及,越来越多的形体通过光学扫描设备输入计算机,形成网格模型,然后借助网格分割技术从中提取有用的结构、特征信息。当前基于曲率的分割方法占据主导地位,但常不能取得满意的分割结果。介绍了一种基于分层特征拟合的网格分割方法,适用于工业产品的反求设计,且可以处理扫描数据残缺的情形。详述了算法思路、流程,并给出了反求工程应用实例。

Freeform shape is usually designed by reverse engineering method thorough a 3D scan-ner,so points cloud segmentation becomes a key problem. Curvature based segmentation algorithms are most used today,and can＇t easily deal with holes and noises in the measurement data.A feature-fitting based segmentation method is introduced,and ideals behind the algorithm are explained,and the computing steps are visualized thorough a simple example.Traditional segmentation algorithm and the new one are compared in the reverse engineering process of a complex automobile part. The result shows that the feature-fitting based segmentation method is more robust and efficient.