当前位置:主页 > 科技论文 > 软件论文 >

基于脊谷线的三维点云孔洞补缺方法研究

发布时间:2018-12-15 07:37
【摘要】:随着三维信息技术的发展,孔洞修补技术成为一项重要的研究课题。但至今对带有尖锐特征或者具有多种曲面类型的复杂曲面的孔洞修补效果仍不尽如人意。本文结合点云物体的显著特征-脊谷特征,给出了基于脊、谷线拟合的孔洞修补方法。主要工作集中在以下几个方面:(1)针对孔洞修补的复杂性问题,给出了一种基于人机交互配对的脊、谷线修复方法。该方法首先根据象限进行边界点的提取,并利用点的曲率特征获取脊、谷特征点;然后分别对其进行分组,采用最小生成树的方法将这些点连接形成边界线和脊、谷线;最后基于人机交互选取孔洞边界处的脊、谷线进行配对,进而采用曲线拟合的方法进行脊、谷线整合,恢复出孔洞上的脊、谷线。(2)针对小孔洞修补的尖锐特征保持问题,给出了一种基于脊、谷线控制的孔洞曲面拟合的点云孔洞整体补缺方法。该方法首先进行孔洞边界点的配对曲线拟合,生成一系列的曲线;然后在表征着孔洞处结构信息的脊、谷线的限制下,对这些拟合曲线进行插值修整,生成控制点序列;最后采用NURBS曲面进行曲面拟合来完成孔洞的修补。(3)针对大孔洞修补的轮廓形状保持问题,设计并实现了一种基于脊、谷线的点云孔洞分块补缺方法。该方法首先基于修复出的脊、谷线进行孔洞的划分;然后对每个划分出的小孔洞,在边界线以及脊、谷线的限制下,计算出每个被划分出来的小孔洞的NURBS曲面拟合的控制点序列,最后对每个小孔洞进行NURBS曲面拟合,最终完成孔洞的修补。本文的方法,不仅为点云测量中不可避免的孔洞修复给予了有效支撑,并为基于点云的各类方法在实测点云上的鲁棒性应用提供了良好途径。
[Abstract]:With the development of three-dimensional information technology, hole repair technology has become an important research topic. However, the effect of hole repair for complex surfaces with sharp features or various surface types is still unsatisfactory. In this paper, the feature of ridge-valley, which is a prominent feature of cloud objects, is combined, and a method of hole repair based on ridge and valley line fitting is presented. The main work focuses on the following aspects: (1) aiming at the complexity of hole repair, a method of ridge and valley line restoration based on human-computer interaction pairing is presented. Firstly, the boundary points are extracted according to the quadrant, and the ridge and valley feature points are obtained by the curvature feature of the points, and then the points are grouped into groups, and the minimum spanning tree is used to connect these points to form boundary lines, ridges and valley lines. Finally, the ridge at the hole boundary is selected based on human-computer interaction, the valley line is matched, and then the ridge and valley line are integrated by curve fitting method to restore the ridge and valley line on the hole. (2) to maintain the sharp feature of small hole repair. In this paper, a new method of hole surface fitting based on ridge and valley line is presented. In this method, a series of curves are generated by fitting the matching curves of the hole boundary points, and then, under the constraints of ridges and valley lines that represent the structural information of the holes, the fitting curves are interpolated to generate control point sequences. Finally, the surface fitting of NURBS surface is used to complete the repair of holes. (3) aiming at the problem of maintaining the contour of large holes, a method of filling holes in point cloud based on ridges and valley lines is designed and implemented. The method is based on the restored ridges and valley lines. Then, under the limit of boundary line, ridge and valley line, the control point sequence of NURBS surface fitting for each divided small hole is calculated. Finally, the NURBS surface fitting for each small hole is carried out. The final completion of the hole repair. The method in this paper not only provides an effective support for the inevitable hole restoration in point cloud measurement, but also provides a good way for the robust application of various methods based on point cloud on the measured point cloud.
【学位授予单位】:西安理工大学
【学位级别】:硕士
【学位授予年份】:2016
【分类号】:TP391.41

【参考文献】

相关期刊论文 前10条

1 何桂珍;;基于特征数据分块自适应切片的空洞修补[J];华东交通大学学报;2014年04期

2 王永波;盛业华;;一种基于曲率极值法的LiDAR点云特征提取算法[J];中国矿业大学学报;2011年04期

3 王小超;曹俊杰;刘秀平;李宝军;;波前法在三角网格孔洞修补中的应用[J];计算机辅助设计与图形学学报;2011年06期

4 李晔;王映辉;刘晶;;基于脊谷特征提取的三维人脸识别[J];计算机工程与应用;2011年12期

5 王英惠;吴维勇;;基于局部细分曲面拟合的散乱数据孔洞修补算法[J];机械设计;2009年12期

6 钟毅;林德静;;基于移动最小二乘法的点云空洞修补算法[J];北京服装学院学报(自然科学版);2008年03期

7 韦争亮;钟约先;袁朝龙;李仁举;;三角网格大面积孔洞光顺修补算法的研究[J];中国机械工程;2008年08期

8 张洁;岳玮宁;王楠;汪国平;;三角网格模型的各向异性孔洞修补算法[J];计算机辅助设计与图形学学报;2007年07期

9 陈飞舟;陈志杨;丁展;叶修梓;张三元;;基于径向基函数的残缺点云数据修复[J];计算机辅助设计与图形学学报;2006年09期

10 王宏涛;张丽艳;李忠文;刘胜兰;周儒荣;;基于RBF神经网络的三角网格曲面孔洞修补[J];中国机械工程;2005年23期

相关硕士学位论文 前3条

1 王珍珍;基于特征增强的三维网格孔洞修补算法研究[D];山东师范大学;2014年

2 王乾;三角网格模型过渡与孔洞修补算法的研究及应用[D];南京航空航天大学;2007年

3 杜佶;径向基函数在逆向工程中的应用研究[D];南京航空航天大学;2005年



本文编号:2380268

资料下载
论文发表

本文链接:https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2380268.html


Copyright(c)文论论文网All Rights Reserved | 网站地图 |

版权申明:资料由用户f7851***提供,本站仅收录摘要或目录,作者需要删除请E-mail邮箱bigeng88@qq.com