一种NURBS曲面拟合精度的量化评测算法

摘要

Surface reconstruction is one of the core technologies in reverse engineering. The advantages of NURBS surface on light smoothing and regional editable characteristics turn it to be a common form of point cloud data freeform surface reconstruction. Nowadays there are some achievements on the study of NURBS surface reconstruction technologies. Since different methods have different fitting accuracies and effects, it is necessary to do research on NURBS surface fitting accuracy evaluation algorithms. Based on. NURBS surface fitting, the paper studies the quantitative indicator of fitting accuracy, designs a search iteration algorithm based on region divisions by which to quickly compute the deviation between primitive point cloud data and NURBS surface.%曲面重构是逆向工程中的核心技术之一,由于NURBS曲面在光顺性和局部可编辑等方面所具有的优点,使其成为点云数据自由曲面重构的常见形式.目前对NURBS曲面重构技术的研究上取得了一些成果,但各方法在拟合精度和效果上各有参差,因此有必要对NURBS曲面拟合精度评价算法进行研究.在采用NURBS实现曲面拟合的基础上,对拟合精度的量化指标进行研究,设计了一种基于区域划分的搜索迭代算法,可快速地计算得到原始点云与NURBS曲面的偏差.