羊体三维激光扫描数据的处理与应用

发布时间：2018-04-10 22:28

本文选题：三维激光扫描技术 + 羊体点云　；参考：《内蒙古农业大学》2017年硕士论文

【摘要】：利用三维激光点云数据,实现实体的三维重构模型具有重要的实用价值,成为测量测绘、图形图像处理等多领域的热点研究内容。羊作为内蒙古地区主要的家畜之一,具有非常重要的经济价值和研究意义,本文以羊体为例,实现了羊体三维激光点云数据的三维重构。利用三维激光扫描仪对目标物体进行采样的过程中,由于设备自身、目标材质,以及扫描过程中手臂的抖动等问题,采集到的点集数据中不可避免的会存在错误点,即噪声数据,噪声数据会影响后期羊体重建精度,造成羊体建模失真等问题;伴随着科研技术的不断进步,三维激光扫描的仪器设备不断升级,采集精度逐步提升,采样的数据量随之变大,数据量过多势必会影响到数据处理速度和存储效率。因此,在对羊体进行点云数据进行三维重构之前,必须对其进行去噪、精简处理。本文以羊体三维激光扫描数据为研究对象,对点云去噪、点云精简、点云三角剖分三方面进行了探讨研究,论文的主要内容如下:(1)针对利用HandyScan 3D手持式三维激光扫描仪采集到的羊体模型点云数据,含有离群噪声数据的情况,提出了结合k-邻域搜索去噪方法和中值滤波方法,实现离群噪声数据的剔除。同时,改进了 k-邻域搜索方法,以此改善点云数据中近邻点的搜索效率。(2)分别利用均匀包围盒方法、曲率精简方法和基于八叉树编码的区域重心方法对羊体点云数据样本进行数据精简处理,并对精简后的数据结果进行分析,总结了各个方法的优缺点,对比得出灵活性更高的基于八叉树编码的区域重心数据精简方法更符合精简要求。(3)介绍了 Crust三角剖分方法,并以Crust三角剖分方法对羊体点云数据进行剖分,得到较好的三角剖分效果,实现羊体三维重构。
[Abstract]:Using 3D laser point cloud data to realize 3D reconstruction model of solid has important practical value and has become a hot research content in many fields, such as surveying and mapping, graphic and image processing and so on.Sheep, as one of the main livestock in Inner Mongolia, has very important economic value and research significance. In this paper, we take sheep body as an example to realize 3D reconstruction of goat body 3D laser point cloud data.In the process of sampling the target object by using 3D laser scanner, due to the problems of equipment itself, target material, and arm jitter during the scanning process, errors will inevitably exist in the collected point set data.That is, noise data and noise data will affect the accuracy of sheep weight building in the later period, resulting in the distortion of sheep body modeling, etc. With the development of scientific research technology, the equipment and equipment of 3D laser scanning are constantly upgraded, and the acquisition accuracy is gradually improved.The amount of data sampled becomes larger, and too much data will affect the speed of data processing and the efficiency of storage.Therefore, before the point cloud data can be reconstructed, it must be de-noised and reduced.In this paper, three aspects of point cloud denoising, point cloud reduction and point cloud triangulation are studied based on the three dimensional laser scanning data of sheep body.The main contents of this paper are as follows: (1) aiming at the point cloud data of sheep body model collected by HandyScan 3D handheld 3D laser scanner and containing outliers noise data, a new method of de-noising based on k-neighborhood search and median filtering method is proposed.The outlier noise data are eliminated.At the same time, the k-neighborhood search method is improved to improve the search efficiency of nearest neighbor points in point cloud data.The curvature reduction method and the region barycenter method based on octree coding are used to simplify the data of sheep body point cloud data, and the results are analyzed, and the advantages and disadvantages of each method are summarized.Compared with the more flexible regional barycenter data reduction method based on octree coding, this paper introduces the Crust triangulation method, and uses Crust triangulation method to divide the goat body point cloud data.A better triangulation effect is obtained and the three-dimensional reconstruction of sheep body is realized.
【学位授予单位】：内蒙古农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41;TN249

【参考文献】