点云的拼接与拓扑-工程力学专业论文.docx

万方数据 万方数据 河北工业大学硕士学位论文 河北工业大学硕士学位论文 摘 要 石油化工设备如果发生泄露问题，往往会引起较大的事故，利用视觉机器人对其 零部件经常性的安全监测是非常必要的。为了使机器人能够快速找到特定零部件，需 要建立石化管道的特定零件数据模型。其中点云模型利用采样点的位置、形状等属性， 通过拓扑连接信息，可以便捷建立三维模型。该模型数据形式简单，易于存储和编辑， 使其在应用中具有很大的优势。 本文利用 CCD 摄像机所获得化工零件目标体的点云信息，对获取的点云信息数 据预处理技术、点云拼接技术以及点云拓扑结构关系进行了深入研究，并建立零部件 的三维数据模型。主要完成了以下几方面工作： 1、将计算机获取的深度图像进行零件的提取操作，得到完整的零件点云信息， 并对其进行去噪处理，建立的 8-邻域结构，进一步通过估算曲率的方法对零件点云进 行简化。 2、提出一种基于公共点改进的 ICP 算法对散乱点云进行拼接，并基于骨架法的 理论建立了点云拓扑关系，以关键点与边界点组合为完整的特征点，对特征点进行边 界拟合，从而获得零件模型形状特征的边界曲线。 3、基于点云簇—簇几何拓扑结构关系和目标体的语义描述方法，建立零部件三 维点云模型库。 关键词：点云数据处理，边界特征提取，点云拼接，点云拓扑 I 点云的拼接与拓扑Skeletal Methodolog 点云的拼接与拓扑 Skeletal Methodology, building point cloud topology relationship. Using the combination of key points and boundary points to create complete special points. Matching them with ABSTRACT If a leak occurs in the petrochemical equipment, it will cause very serious problems. Hence it’s quite necessary to check the safety of petrochemical equipment regularly by using robot vision system. In order to make sure that robots can quickly find certain component, there is a need to build the data model of special parts of petrochemical pipelines. Point Cloud Model is a way to attach this goal. It can easily build three-dimensionalmodel through linking information (location, shape, etc.) of sampling points. This model data has a simple structure, which is easy to Storage and editing, making it have the great advantage in the application. In this assignment, CCD camera is used to gather point cloud data of petrochemical components. There is a depth study on the pretreatment technology, point cloud splicing technology and point cloud topology relations of gathered data.Besides that, a three-dimensionaldata model of components is also built in this study. Followed are the main jobs that are going to be finished in this assignment: Conducting extraction operation of components on the depth-image that is shown on the computer to get complete point cloud information. After that there is a need to do