本文關(guān)鍵詞：基于分割點(diǎn)云的NURBS曲面三維重構(gòu)方法研究　出處：《北京建筑大學(xué)》2013年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 三維激光 NURBS 點(diǎn)云切片 曲面重構(gòu) OpenGL

【摘要】：三維激光掃描作為一種高速收集對象精細(xì)、稠密的表面數(shù)據(jù)的掃描技術(shù)，近年來在工業(yè)設(shè)計(jì)、現(xiàn)代大型異構(gòu)工程施工、文物保護(hù)等領(lǐng)域得到了廣泛的使用。以上領(lǐng)域都需要對物體的原始點(diǎn)云數(shù)據(jù)進(jìn)行三維構(gòu)建，在構(gòu)建結(jié)果的基礎(chǔ)上分析和利用。例如：在大型異構(gòu)工程構(gòu)件的施工放樣中，需要對構(gòu)件重建，在計(jì)算機(jī)中模擬其拼接效果，并根據(jù)需要實(shí)時(shí)對其姿態(tài)等進(jìn)行調(diào)整；在文物保護(hù)與文化遺產(chǎn)領(lǐng)域，需要重構(gòu)掃描物體的外形特征，以便現(xiàn)實(shí)中重建、分析、仿真等應(yīng)用；在工業(yè)設(shè)計(jì)中，一些物體細(xì)節(jié)處的重要特征很難獲取，但通過重建物體表面可以輕易的將手工雕刻模型輸入到CAD系統(tǒng)中；在逆向工程中，往往有大量的傳統(tǒng)零部件必須先測量重建出其表面形狀信息進(jìn)而改進(jìn)修正設(shè)計(jì)出的新品；另外，重建也可以作為檢查站站之間數(shù)據(jù)遺漏或空洞的重要手段。 基于點(diǎn)云的三維重建在整個(gè)數(shù)字化過程中起著關(guān)鍵的作用，重構(gòu)模型的精度與速度直接影響后續(xù)工作的開展。因此尋找一種高效靈活的建模方法是十分必要的。本文綜合闡述了基于NURBS理論進(jìn)行曲面重構(gòu)的原理、研究現(xiàn)狀和技術(shù)方案，并分析了該方向的總體發(fā)展和現(xiàn)有解決方案的不足之處，得出基于切片技術(shù)構(gòu)建分割點(diǎn)云的NURBS曲面的方法。論文的主要研究內(nèi)容包括下面四部分： 1．對已分割的散亂點(diǎn)云數(shù)據(jù)采用點(diǎn)云切片技術(shù)，提取用于構(gòu)建NURBS曲線曲面的型值點(diǎn)序列，并研究了點(diǎn)云切片的參數(shù)設(shè)置規(guī)則。這些參數(shù)包括切片的層數(shù)，切片的行進(jìn)路線，切片之間的距離以及切片本身的厚薄程度，并深入研究了對于不同特征的點(diǎn)云的切片行進(jìn)路線的判定方法。 2．在構(gòu)建NURBS之前，對切片數(shù)據(jù)進(jìn)行預(yù)處理，采用一種逐點(diǎn)斜率求差法進(jìn)行數(shù)據(jù)壓縮，，用最少的數(shù)據(jù)點(diǎn)保證曲面的形狀，根據(jù)處理后得到的新數(shù)據(jù)點(diǎn)求取控制點(diǎn)，也可以達(dá)到用存儲量小的控制點(diǎn)重構(gòu)符合精度要求的NURBS曲面。 3．在計(jì)算NURBS構(gòu)建所需要的控制點(diǎn)時(shí)，本文選用三次B樣條曲線，采用重節(jié)點(diǎn)技術(shù)進(jìn)行曲線曲面插值，其中的附加條件采用以首末型值點(diǎn)為基礎(chǔ)，通過方程解算控制點(diǎn)，是一種不同于邊界附加條件的簡單控制頂點(diǎn)解算方法。 4．對構(gòu)建的NURBS曲面采用基于網(wǎng)格劃分的迭代搜索法進(jìn)行精度分析。以點(diǎn)云到曲面的相關(guān)距離數(shù)值為依據(jù)進(jìn)行評定。應(yīng)用這種方法，可以減小搜索范圍，提高搜索的效率。 根據(jù)本文研究內(nèi)容的具體含義和選擇的技術(shù)方案，即以分割點(diǎn)云數(shù)據(jù)為基礎(chǔ)，以構(gòu)建雙三次NURBS曲面為目的，采用快捷、簡單、精確的方法進(jìn)行三維重構(gòu)，并對構(gòu)建的曲面進(jìn)行精度分析。本文使用C++高級編程語言，結(jié)合OpenGL開發(fā)實(shí)驗(yàn)系統(tǒng)，選用項(xiàng)目中的相關(guān)數(shù)據(jù)來驗(yàn)證算法的可實(shí)用性，總結(jié)本文研究的技術(shù)路線，以及其實(shí)用性與精確性等。
[Abstract]:3D laser scanning as a high speed collection of fine, dense surface data scanning technology, in recent years in industrial design, modern large-scale heterogeneous engineering construction. Cultural relic conservation and other fields have been widely used. The above fields need to construct the original point cloud data. On the basis of the results of the construction analysis and utilization, for example, in the construction of large heterogeneous engineering components, it is necessary to rebuild the components, and simulate the splicing effect in the computer. And according to the need of real-time adjustment of its attitude; In the field of cultural relics protection and cultural heritage, it is necessary to reconstruct the external features of scanned objects for the purpose of reconstruction, analysis, simulation and other applications in reality. In industrial design, it is difficult to obtain the important features of some object details, but the manual carving model can be easily input into the CAD system by reconstructing the surface of the object. In reverse engineering, a large number of traditional parts must first measure and reconstruct their surface shape information to improve the design of new products. In addition, reconstruction can also be used as an important means of missing data or voids between checkpoints. 3D reconstruction based on point cloud plays a key role in the whole digitization process. The precision and speed of the reconstruction model directly affect the development of subsequent work. Therefore, it is necessary to find an efficient and flexible modeling method. This paper comprehensively describes the principle of surface reconstruction based on NURBS theory. The current situation and technical scheme are studied, and the overall development of this direction and the shortcomings of the existing solutions are analyzed. The method of constructing NURBS surface based on slicing technology is obtained. The main contents of this paper include the following four parts: 1. Point cloud slicing technique is used to extract the sequence of value points used to construct NURBS curve and surface. These parameters include the number of layers, the route of the slice, the distance between slices and the thickness and thickness of the slice itself. The method of determining the path of slicing of point clouds with different characteristics is also studied in depth. 2. Before constructing NURBS, the slice data is preprocessed, a method of point by point slope difference is used to compress the data, and the shape of the surface is guaranteed with the least number of data points. According to the new data points obtained after processing, the control points can be obtained, and the NURBS surface can be reconstructed by using the control points with small storage capacity to meet the precision requirements. 3. In the calculation of the control points needed by NURBS, the cubic B-spline curve is selected, and the curve and surface interpolation is carried out by using the heavy node technique, in which the additional conditions are based on the first and last value points. The solution of control points by equation is a simple method for solving control vertices, which is different from the boundary additional conditions. 4. The precision of the constructed NURBS surface is analyzed by the iterative search method based on mesh division. Based on the relative distance between the point cloud and the surface, the method is applied. Can reduce the search range, improve the efficiency of the search. According to the specific meaning of the research content and the choice of the technical scheme, that is, based on the segmentation of point cloud data, to build a bicubic NURBS surface for the purpose of using fast and simple. In this paper, we use C high-level programming language, combined with OpenGL to develop the experimental system. The relevant data in the project are selected to verify the practicability of the algorithm, and the technical route, practicability and accuracy of this paper are summarized.
【學(xué)位授予單位】：北京建筑大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2013
【分類號】：P225.2

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王飛;三次B樣條反算的一種簡便算法[J];北京郵電大學(xué)學(xué)報(bào);1996年03期

2 王晏民;矢量曲線的特征點(diǎn)提取[J];測繪工程;2002年02期

3 朱根松;周天瑞;;反求工程中任意平面切片算法及曲線快速重構(gòu)[J];鍛壓技術(shù);2008年03期

4 張勝文;張洪剛;方喜峰;李敏;;基于切片技術(shù)的點(diǎn)云數(shù)據(jù)預(yù)處理研究[J];江蘇科技大學(xué)學(xué)報(bào)(自然科學(xué)版);2009年05期

5 王晏民,李德仁;多函數(shù)分段擬合的曲線數(shù)據(jù)壓縮[J];哈爾濱工業(yè)大學(xué)學(xué)報(bào);2002年04期

6 李帥;方源敏;喜文飛;;矢量曲線特征點(diǎn)提取方法的改進(jìn)[J];河南科學(xué);2011年04期

7 葉麗;謝明紅;;采用積累弦長法擬合3次NURBS曲線[J];華僑大學(xué)學(xué)報(bào)(自然科學(xué)版);2010年04期

8 劉煥彬;余正生;;點(diǎn)云管道曲面的重構(gòu)[J];杭州電子科技大學(xué)學(xué)報(bào);2011年02期

9 危雙豐;王晏民;;一種深度圖像擬合圓柱的優(yōu)化方法[J];測繪科學(xué)技術(shù)學(xué)報(bào);2007年05期

10 劉云峰,柯映林;反求工程中的混合切片技術(shù)[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2003年06期

相關(guān)博士學(xué)位論文 前3條

1 呂震;反求工程CAD建模中的特征技術(shù)研究[D];浙江大學(xué);2002年

2 劉云峰;基于截面特征的反求工程CAD建模關(guān)鍵技術(shù)研究[D];浙江大學(xué);2004年

3 孟娜;基于激光掃描點(diǎn)云的數(shù)據(jù)處理技術(shù)研究[D];山東大學(xué);2009年

相關(guān)碩士學(xué)位論文 前6條

1 劉丹;基于點(diǎn)云數(shù)據(jù)的曲線擬合[D];遼寧工程技術(shù)大學(xué);2009年

2 劉煥彬;點(diǎn)云管道曲面重構(gòu)[D];杭州電子科技大學(xué);2011年

3 范宜艷;基于NURBS的自由曲面重構(gòu)技術(shù)研究及開發(fā)[D];天津大學(xué);2004年

4 朱海軍;復(fù)雜曲面反求技術(shù)的研究[D];南京理工大學(xué);2005年

5 楊憲武;車身曲面重構(gòu)中三維散亂點(diǎn)云的參數(shù)化及擬合[D];山東理工大學(xué);2007年

6 余國鑫;逆向工程曲面重建技術(shù)的研究與應(yīng)用[D];廣東工業(yè)大學(xué);2008年

本文編號：1437561