摘要
针对静态爆轰试验后发生形变的靶板,提出一种新的参数化方法,在三角网格模型下解决了靶板上孔洞部分参数化形变大以及边界角度形变、面积形变和拉伸形变大的问题。该方法首先通过拟合变分隐式曲面修补孔洞,接着采用基于虚拟边界的ABF++方法进行参数化,然后根据试验前靶板的网格能量,优化整体的靶板点云,进一步减少形变,最后去除附加的修补点和虚拟边界点,得到参数化后的靶板点云。实验选取了LSCM、ABF++方法来直接对两个经典人脸三角网格模型和靶板三角网格模型及其对应带孔洞的模型进行形变恢复,并与所提方法进行了对比,实验结果表明所提方法在三个模型的外边界角度畸变、外边界面积畸变,含孔洞三个模型的内边界角度畸变、内边界面积畸变,四项指标上相较于LSCM和ABF++平均降低了31.10%、23.76%、19.60%、10.08%,因此可以更好地减少参数化曲面上的内外边界形变,为带破片穿孔的靶板提供一种有效的参数化方法。
A new parametric method was proposed for addressing deformations in target plates after static detonation experiments within the framework of a triangular mesh model.This method addressed the significant deformations in the parametric representation of hole regions on the target plate,as well as the distortions in boundary angles,area,and stretching caused by the triangular mesh.The approach begins by fitting variational implicit surfaces to repair the holes,followed by the application of the Virtual Boundary-Based ABF++method for parameterization.Subsequently,it optimized the overall target plate point cloud based on the mesh energy of the target plate before the experiment,further reducing deformations.Finally,it removed additional repair points and virtual boundary points to obtain the parametrically represented target plate point cloud.In experiments,LSCM and ABF++methods were chosen to directly perform deformation recovery on two classic triangular mesh models,namely,the human face model and the target plate model with holes.These results were compared to our proposed method.The experimental results demonstrated that our method outperformed LSCM and ABF++by reducing average distortions in four key metrics:external boundary angle distortion,external boundary area distortion,internal boundary angle distortion,and internal boundary area distortion,by an average of 31.10%,23.76%,19.60%,and 10.08%,respectively.Thus,our approach provided an effective parametric method for reducing deformations on both internal and external boundaries of parametric surfaces,making it a valuable tool for perforated target plates.
作者
张艺驰
高俊钗
张晓倩
ZHANG Yichi;GAO Junchai;ZHANG Xiaoqian(School of Electronic and Information Engineering,Xi'an Technological University,Xi'an 710021,China)
出处
《探测与控制学报》
CSCD
北大核心
2024年第5期34-42,共9页
Journal of Detection & Control
基金
国家自然科学基金项目(62001365)。
关键词
三角网格参数化
虚拟边界
变分隐式曲面
基于角度平展
靶板形变恢复
triangular mesh parameterization
conformal mapping
virtual boundary
variational implicit surface
angle flattening-based
target plate deformation recovery