牙模数据拼合重建系统的研究与实现

上传人:E**** 文档编号:118412945 上传时间:2019-12-15 格式:PDF 页数:77 大小:1.82MB
返回 下载 相关 举报
牙模数据拼合重建系统的研究与实现_第1页
第1页 / 共77页
牙模数据拼合重建系统的研究与实现_第2页
第2页 / 共77页
牙模数据拼合重建系统的研究与实现_第3页
第3页 / 共77页
牙模数据拼合重建系统的研究与实现_第4页
第4页 / 共77页
牙模数据拼合重建系统的研究与实现_第5页
第5页 / 共77页
点击查看更多>>
资源描述

《牙模数据拼合重建系统的研究与实现》由会员分享,可在线阅读,更多相关《牙模数据拼合重建系统的研究与实现(77页珍藏版)》请在金锄头文库上搜索。

1、南京航空航天大学 硕士学位论文 牙模数据拼合重建系统的研究与实现 姓名:郭美社 申请学位级别:硕士 专业:航空宇航制造工程 指导教师:程筱胜 20090101 南京航空航天大学硕士学位论文 i 摘 要 非接触式光学三维测量技术具有非接触、速度快、精度高等特点,近年来在机械设计与制 造、航空航天、生物医学工程、计算机视觉与虚拟现实、文物修复等领域的应用越来越广泛。 牙模具有外形复杂且体积小等特点,因此针对牙模的非接触式光学三维测量系统的研究和开发 具有重要的理论意义和工程应用价值。 非接触式光学三维测量系统主要有两个部分组成:一是被测物体的表面数据的获取;二是 曲面重建。本文的研究工作主要有光学

2、三维测量系统的控制系统、多视点云拼接和以三角网格 为基础的曲面重建技术,具体内容涵盖: (1) 设计光学三维测量系统精确稳定的测量转台。主要包括光栅投影模块和电动旋转台控 制模块。计算机控制投影仪向被测物体投射 4 幅光栅条纹,相机采集经过物体表面高 度调制的变形条纹图像转化为三维坐标数据,旋转台在步进电机的控制下,获得被测 物体的多个不同视角的数据点云, 测量一周后完成物体表面轮廓的 360三维全场测量。 (2) 研究点云噪声的去除和多视点云的拼接,用手动选取三个参考点的方法实现点云的预 拼接和 ICP 算法实现多视点云的精确拼接。 (3) 用二维 Delaunay 三角剖分实现空间散乱点集

3、的三角网格化,该方法克服了三维 Delaunay 三角剖分高复杂度和投影 Delaunay 三角网格算法质量难以保证的缺点, 研究 通过对空间散乱数据进行适当分区,提高 k 近邻搜索算法的速度。 (4) 研究了 VTK 可视化类库和 HOOPS 3D 图形应用接口与 Microsoft Visual Studio .NET 2003 的系统整合。 (5) 开发了三维重建软件原型系统,并用该系统对牙模的点云数据进行了拼合和重建,取 得了很好的效果。 关键字:三维测量,曲面重建,点云拼接,Delaunay 三角剖分,散乱数据 牙模数据拼合重建系统的研究与实现 ii ABSTRACT Non-con

4、tact optical 3D measurement technology has the advantages of non-contact, fast speed and high precision in measuring process, so it has been applied more and more extensively in machine design and manufacture field, aerospace industry, biomedical engineering, computer vision, virtual reality and cul

5、tural relic restoration recently. Teeth model is a kind of object with complicated surface and small volume. Hence the research and develop on a set of Non-contact optical measuring system for dental model means a lot on both theoretical and engineering application. Optical 3D measurement system inc

6、ludes two main aspects: One is surface data acquisition of measured object, the other is surface reconstruction. This thesis includes control system of optical 3D measurement device, multi-view point cloud registration, and surface reconstruction technology based triangular mesh. The concrete conten

7、t of this research includes five subtopics: (1) An accurate and stable motorized rotary stage of optical 3D measurement device is developed。It mainly includes grating projection module and rotary electrical servo. After measured object is perfectly fit on the motorized rotary stage driven by the ste

8、p motor, the computer controls projector to project four grating fringes on it, then transforms the deformed fringe images modulated by the objects height into 3D coordinate data. Motorized rotary stage is driven by the step motor and the multi-view point clouds of measured object can be obtained. A

9、fter measuring around the object with one circle, 360-deg 3D full-court measurement of objects surface profiles will be completed. (2) Study on unorganized point cloud denoising method and multi-view point cloud registration. The method using manual selection of three points is realized pre-registra

10、tion of point cloud data and precise registration based on ICP algorithm. (3) 3D Delaunay neighbors of any points are searched by the map from 2D triangulation to space unorganized points, which solves high complexity of 3D Delaunay triangulation and shortcomings of quality not guaranteed of project

11、ion Delaunay algorithm. This thesis increases searching speed of k-nearest neighbor by proper partition of space unorganized point cloud data. (4) Study on system integration of VTK(Visualization Toolkit) ,HOOPS 3D Application Framework and Microsoft Visual Studio .NET. (5) 3D reconstruction softwar

12、e prototype system has been established, and the point cloud of detal models have been aqured by this system. after sequential merging and reconstruction, the result 南京航空航天大学硕士学位论文 iii shows that the system has a good performance. Keywords: 3D measurement, Surface reconstruction, Point cloud registration, Delaunay triangulation, unorganized point cloud 牙模数据拼合重建系统的研究与实现 vi 图、表清单 图 1.1 快速制造.2 图 1.2 口腔修复.2 图 1.3 文物保护.2 图 1.4 2D Crust 算法.8 图 1.5 Power 图 .8 图 1.6 本文的研究路线.11 图 2.1 VTK 可视化管道.14 图 2 . 2 VTK 流水线框架.14 图 2.3 VTK 数据类型.15 图 2.4 VTK 五大数据类型.

展开阅读全文
相关资源
相关搜索

当前位置:首页 > 学术论文 > 其它学术论文

电脑版 |金锄头文库版权所有
经营许可证:蜀ICP备13022795号 | 川公网安备 51140202000112号