高级检索
胡靖尘, 郑国磊. CAD曲面等值分割方法[J]. 计算机辅助设计与图形学学报, 2021, 33(1): 153-160. DOI: 10.3724/SP.J.1089.2021.18375
引用本文: 胡靖尘, 郑国磊. CAD曲面等值分割方法[J]. 计算机辅助设计与图形学学报, 2021, 33(1): 153-160. DOI: 10.3724/SP.J.1089.2021.18375
Hu Jingchen, Zheng Guolei. CAD Surface Iso-Segmentation[J]. Journal of Computer-Aided Design & Computer Graphics, 2021, 33(1): 153-160. DOI: 10.3724/SP.J.1089.2021.18375
Citation: Hu Jingchen, Zheng Guolei. CAD Surface Iso-Segmentation[J]. Journal of Computer-Aided Design & Computer Graphics, 2021, 33(1): 153-160. DOI: 10.3724/SP.J.1089.2021.18375

CAD曲面等值分割方法

CAD Surface Iso-Segmentation

  • 摘要: CAD曲面分割在工程领域中具有广泛的应用.已有分割方法主要根据单一的面点属性进行分割,不具通用性.为此,提出以自定义条件为核心的面等值分割方法.用户预先根据分割问题及相关面点属性设计一组条件,以此为依据计算界点,并沿界边生长方向依次搜索界点,进而以有序界点样条插值的方式生成界边.通过构造曲面及其子面片的界边,将曲面分割为一组等值面.为保证分割的一般性,提出初始界点和参数域采样单元界点精确计算方法.最后,针对可加工性、法向量和曲率3种面点属性分别设计分割条件并进行实例测试,在可加工区域分割测试中文中方法的平均界点误差比现有方法平均降低96.98%.实验结果表明,所提方法适用于不同的分割问题,且具有较高的分割精度.

     

    Abstract: CAD surface segmentation plays a significant role in engineering fields.The research to date has mainly performed segmentation based on a single attribute,thus suffering from little applicability to different segmentation situations.We present a surface iso-segmentation method focusing on user-defined conditions designed beforehand according to the segmentation problem and relevant attributes.Based on user-defined conditions,the boundary point(BP)is calculated and searched in the growing direction of a boundary curve(BC),and then used for constructing the BC by spline interpolation.By constructing the BC of the input surface and its sub-surfaces,the input surface can be divided into several iso-surfaces.To ensure the generality of segmentation,a precise calculation method for the initial BP and the BP of sampling units on the parameter domain is explored.Finally,attributes including machinability,normal and curvature are used to design conditions for the iso-segmentation of several CAD surfaces,and in the machinable region segmentation test,the average BP deviation of our algorithm is 96.98%lower than that of an existing algorithm.Test results show that the proposed algorithm suits for various segmentation problems,and can achieve more precise segmentation.

     

/

返回文章
返回