Data Reduction Methods for Reverse Engineering

Reverse engineering is an emerging technology that promises to play a role in reducing product development time. Reverse engineering in this paper refers to the process of creating engineering design data from existing parts. It creates or clones an existing part by acquiring its surface data using a scanning or measurement device. Recently, laser scanning technology has improved significantly and has become a viable option in capturing geometries of complicated design models. 3D laser scanners have become more accurate and the speed of data acquisition has increased dramatically. However, they generate up to thousands of points per second, and handling the huge amount of point data is a major problem. It becomes quite important, therefore, to reduce the amount of acquired point data and convert it into formats required by the manufacturing processes while maintaining the accuracy. This paper presents methods that can reduce efficiently the amount of point data.     

基于单双幅图曲线曲面特征点的透视反求及拟合

透视反求是为了从图片获取一定的信息,从而建立物体的三维模型。利用透视图由世界坐标系经过平移、旋转和缩放基本矩阵变换到观察坐标系的原理,推导出单幅透视图和双幅透视图透视反求的方法,分别用这2种基本方法反求出透视图片中特征点的坐标,然后再用这些特征点的坐标来完成重建物体的三维模型。并以单幅透视图和双幅透视图中曲线、曲面特征点为例进行了Bezier曲线、曲面的拟合。利用C++平台的OpenGL(开放性计算机图形库)库得到的试验结果表明该算法实际吻合情况较好。     

层去图像法反求工程中的数据处理

提出层去图像法反求工程中的数据处理方法。首先采用多帧平均、中值滤波和保持细节平滑滤波对图像进行预处理,使图像质量得到恢复与改善,然后再进行图像分割和边缘检测,最后进行轮廓追踪有其多边形逼近。实践表明,所提出的数据处理方法具有较好的应用效果。     

曲线曲面拟合在逆向造型设计中的应用

由Tritop和Atos软件得到汽车车门点云数据，利用CATIA软件的数字编辑器和自由造型单元，对车门点云进行分网和曲面拟合处理，得到了车门曲面造型的结果。在处理过程中，运用了编辑点云、划界、过滤、分网、分区拟合和链接等一系列方法，得出了一些应用CATIA软件的经验和技巧。     

一种新的散乱数据边界点提取方法

综合采用了邻域和平面上提取边界点的思想 ,提出了一种新的在二维平面上提取边界点的方法 ;并在曲面局部近似一个平面的前提下把它推广到三维空间 ,从而形成了一种可直接在三维空间对任意复杂形状曲面进行边界点提取的 3D方法     

逆向工程中点云处理及拟合新方法的研究

基于逆向工程软件imageware的设计基本思路,以某轧辊的点云数据作为研究对象,对点云处理及拟合方法进行对比分析。结果表明:基于点云曲率拟合基准平面能够显著减小拟合误差,提高拟合精度;B样条曲线拟合不仅拟合精度较高,而且曲线光顺性较好;基于B样条的分段拟合方法能够解决B样条曲线不能精确反映二次曲线的问题,提高拟合精度;随着截面数量的增加,曲面-点云偏差减小,当截面数量增加到一定程度时,曲面-点云偏差缓慢增加,最后趋于稳定。     

逆向工程中二次曲面拟合方法的研究

数据分块和曲面拟合是逆向工程中最关键的部分。主要探讨并实现了一种新的二次曲面拟合算法：首先用一种改进的参数化方法建立球面、圆柱面、圆锥面和圆环面的几何距离函数，然后用非线性最小二乘法拟合曲面。实验证明所采用的拟合方法能够有效避免奇异值问题，具有较好的健壮性和拟合精度较好等优点。     

Segmentation of the Measured Point Data in Reverse Engineering

Laser scanning is widely used because of its fast measuring capabilities and high precision. The segmentation of the scanned data is necessary for the fast and efficient surface modelling, but most segmentation techniques are based on very regular data, and the adaptation of previous techniques to process the scanned data does not usually produce a good result. In this paper, the solution of the segmentation of noise-free data and noisy data is proposed and compared. The solution for noise-free data is relatively simple having a small number of user-defined criteria. However, the solution for noisy data requires more criteria to produce a reasonable result when noise is globally distributed on the measured data. The approach is based on triangulated data and the result depends on some user-defined criteria. The result is illustrated to demonstrate its adaptability to measured data on a freeform surface and the results by the different criteria are compared. ID="A1" Correspondence and offprint requests to: Dr H.-C. Kim, Production and Automation Laboratory, Department of Mechanical and Intelligent Systems Engineering, Pusan National University, KeumJeon Gu, Pusan 609–735, Korea. E-mail: hchakim@hyowon.pusan.ac.kr     

A Freeform Surface Modelling System Based on Laser Scan Data for Reverse Engineering

Reverse engineering technology starts with a measurement process, inspecting the surface contour of the object, and applies a CAD geometric modelling technique to construct the surface model. The purpose of this study is to develop a freeform surface modelling system involving the functions of data reorganisation and surface construction of reverse engineering. An important task in reverse engineering is the process of reducing the large amount of coordinate data derived by measurement devices. Reduction methods in common use include uniform deviation, division by a factor, spatial sampling, chordal deviation and interpolated sampling. Each of the above modes has its advantages and weaknesses, and suitable application areas. A new data reduction mode, which integrates the space concept and height characteristic, is proposed, called the height decision method. This method can achieve the aim of data simplification while maintaining the curve characteristics of the model. The CAD modelling system is developed by an AutoLISP program with MDT v4.0 software. An egg model is presented in this paper for explanation and verification.     

基于逆向工程的车门外侧板设计

主要讨论了利用逆向工程技术进行车身覆盖件点云数据处理以及曲面造型过程中的基本要点及关键技术。事实证明,逆向技术可以大大减少产品的开发周期,最大限度地体现造型设计师的设计风格。     

逆向工程中的数据精简研究

逆向工程中,数据精简技术是进行曲面重构的重要内容。介绍了曲面曲率的基本理论,根据曲率精简点云的方法,并结合实际对根据曲率精简点云数据的方法进行改进,着重研究了邻域搜索、曲率计算、曲率精简原则等内容,在保证测量精度的前提下,时提高曲面重构的计算效率、减少存储空间等有较好的效果。     

反向工程数据获取和几何建模方法

反向工程是一种快速的产品设计、制造方法,是根据实物或模型测得的数据,构造 ＣＡＤ模型,继而进行分析制造。本文总结了反向工程数据获取和几何建模的现有方法, 指出了各种方法的优缺点,提出了反向工程几何建模今后的发展方向及研究重点。     

G2 Continuity for Multiple Surfaces Fitting

Fitting 3D known data into sculptured surfaces has received extensive attention. However, most algorithms available are suitable only for an isolated surface. A part typically contains multiple surface regions that must be blended to a degree of continuity. The conventional approach based on blending, lofting, stitching, etc. cannot guarantee the quality of the surface near the jointed area. The purpose of this work is to present a surface-fitting algorithm for multiple sets of data, concentrating on G 2 continuity across the boundary of the fitted surfaces. The proposed surface-fitting algorithm essentially fits several sets of data simultaneously, and yields a B-spline surface for each set of data. The G 0 ,G 1 , and G 2 continuity conditions between B-spline surface patches were addressed. Based on these results, additional constraints were specified to achieve G 2 continuity across the surface boundary. A successful demonstration of the proposed strategy is provided also.     

基于逆向工程技术的后牙嵌体的数字化个性设计

以逆向工程技术为基础,对牙合嵌体的数字化个性设计展开探讨根据各部分结构特点和功能要求,以三角网格代替曲面拼接,反求组织面,保证了精度,提高了效率;以基于特征线的UV曲面重构牙合面、轴面,恢复了生理特征。对密合度这个关注焦点,本文以边缘线作为各部分造型的边界约束,实现了无缝配合。设计出的嵌体不仅符合生理解剖的一般规律,也满足了个性特征及咬合规律,达到临床医学要求。     

一种逼近（过）型值点的曲线匀化方法

本文提出一种在工程上实用的Ｂｅｚｉｅｒ曲线匀化方法。在一定条件下，匀化曲线可按照指定精度逼近（过）原曲线的型值点列。文中首先介绍了基于离散与归并方法的匀化条件及实现步骤。对过节（型值）点的曲线归并方法做了进一步阐述。文中所述内容已在ＮＰＵ－１／ＣＡＤ－ＣＡＭ系统软件中实现和应用。     

精度设计——逆向工程中反求设计的关键一环

讨论了精度设计在逆向工程中的作用、特点及其实现方法。阐述了反求设计的三个主要组成部分 :三维重构、原始设计参数还原和精度设计的作用及其相互关系。详细论述了封闭环的选择与公差确认在反求设计中的特殊性 ,并提出对于各项精度指标的判断与决断是区分正、反向设计的重要标志。最后 ,总结提出了逆向工程中精度设计的实现步骤。     

基于插补算法的数据拟合与曲线显示

根据点阵液晶显示模块显示点地址的特点,对点阵液晶屏幕进行了规划,推导出显示点地址与(x,y)坐标点之间的映射关系。在对曲线进行分段拟合的逐点比较法分析的基础上,提出了用“水平、垂直、±45°方向”进行曲线数据拟合的插补算法。论述了其插补的基本原理,给出了插补算法表,用简单的算法实现曲线和实时采集数据曲线的拟合与显示。     

逆向工程中自由曲面自组织重建研究

探讨了曲面密集三维散乱点数据的自由曲面自组织重建方法。建立了基于扩展自组织特征映射神经网络的自由曲面自组织重建模型及其训练算法。所建模型的网络神经元对曲面散乱点的学习来模拟曲面上的点与点之间的内在关系 ,神经元连接权矢量集重构曲面样本点的内在拓扑关系。经过训练 ,神经网络将整个曲面散乱点数据分成许多子区域 ,子区域的分类核心即为神经元连接权矢量 ,每个子区域用一个线性函数逼近 ,实现自由曲面自组织重建。计算机仿真表明 ,所建神经网络模型可实现三维密集散乱点数据自组织压缩及曲面自组织重建于一体。     

逆向工程中的多视数据对齐方法

在分析了多视数据产生的原因及其不同的数据采集方法,给出了解决多视数据对齐问题的方法。并根据三基准点对齐的原理,采用合适的对齐方法,通过变换坐标从而达到数据对齐定位的目的。本文介绍了目前在逆向工程中常用的几种数据对齐方法。     

基于测量点的草图平面与草图轮廓生成算法研究

提出了在 CATIA二次开发平台上 ,首先由测量点生成草图轮廓 ,再根据用户对设计意图的理解 ,按照正向设计的思路 ,反求结构件三维模型的思想。主要研究了基于测量点的草图平面的提取 ,草图平面上草图轮廓线生成的相关算法。并提出利用类似草图轮廓生成的算法 ,由测量点构造曲面上的若干截面线 ,并构造出曲面上的特征线 ,然后反求自由曲面模型的方法。