基于动态误差控制和PSO的三角网格模型简化优化方法  

Simplification and optimization of triangular mesh model based on dynamic error control and PSO

作　　者：王武礼[1,2,3] 段黎明[1,2] 王浩宇[1,2] 白洋[1,2] 

机构地区：[1]重庆大学光电技术及系统教育部重点实验室ICT研究中心,重庆400044 [2]重庆大学机械工程学院,重庆400044 [3]中国石油大学(华东)信息与控制工程学院,山东青岛266580

出　　处：《计算机集成制造系统》2018年第1期63-71,共9页Computer Integrated Manufacturing Systems

基　　金：国家重大科学仪器设备开发专项资助项目(2013YQ030629).

摘　　要：针对基于工业CT切片重建的三角网格模型简化时,大多数简化算法无法同时兼顾网格的细节特征、简化比例和网格质量等问题,提出一种基于动态误差控制和粒子群优化算法相结合的三角网格模型简化优化方法。该方法分为网格简化和网格优化两个阶段。首先,采用投影预测的方法确定折叠点,使用动态变化的距离误差阈值和角度误差阈值计算三角形的折叠代价,并按照折叠代价从小到大的顺序简化网格模型;然后,利用PSO算法对简化模型中的狭长三角形进行优化处理。实验结果表明,与现有网格简化方法相比,所提方法在控制网格模型简化比例的同时,可有效保持网格细节特征,生成高质量的简化网格模型。Most of the existing mesh simplification algorithms cannot take into account the details,reduction ratio and mesh quality at the same time.To solve the problem,an adaptive simplification and optimization method for triangular mesh based on dynamic error control and Particle Swarm Optimization（PSO）was proposed.The method was distributed into two stages：mesh simplification and mesh optimization.The projection estimation method was used to determine the collapsing vertex,and both threshold of dynamic distance error and threshold of dynamic angle error were introduced to calculate the collapsing cost,and the mesh was simplified according to the cost.PSO was employed to standardize narrow and long triangles.Compared with the existing methods,the experimental results showed that the proposed method could effectively control reduction ratio of mesh and produce high-quality simplified mesh while preserving the details of original model.

关 键 词：逆向工程 网格简化 网格优化 粒子群优化算法 动态误差阈值 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]