面向无线体域网的数字化3D人体模型及其简化

发布时间：2018-03-27 14:04

本文选题：无线体域网(WBAN)　切入点：点云数据　出处：《西安电子科技大学》2014年硕士论文

【摘要】：近年来,随着传感器技术和无线通信技术的发展,以人体为中心,由各种用于监测人体生理指标信号的传感器节点组成的无线体域网(Wireless Body Area Network,WBAN)成为人们关注的焦点。无线体域网作为一种新的无线通信网络,具有与其它无线通信网络不同的特点和要求。其微型性和实用性使得它在医疗诊断、健康监护和消费电子等领域具有广阔的应用前景。目前对无线体域网的研究还处于起步阶段,离技术成熟和广泛应用还有很大一段距离,面临的挑战还很多,还有很多关键性的技术问题有待进一步的研究和解决。无线体域网是架设在人体身上的无线通信网络,人体作为组建网络的传感器节点的主要承载体,是无线体域网的研究工作中必不可少的一部分。由于利用真实的物理人体体模进行相应的研究实验缺乏一定的安全性,而且成本高、效率低,这就需要一种能够取代真实的物理人体体模在无线体域网中的角色的东西——数字化3D人体模型。数字化3D人体模型也称为计算机人,它是通过计算机技术和图像处理技术虚拟出来的能够反应出人体三维形貌特征的3D模型。这种3D人体模型能够在计算机中存储和可视化,并且能够任意调节和控制,相比于真实的物理人体体模,它更加安全、方便和灵活。因此它可以用来进行无线体域网的各种实验仿真,极大地方便了无线体域网的研究,在某种程度上促进了无线体域网的发展。本文首先通过3D激光扫描仪对人体进行扫描测量,获取人体的三维点云数据,接着对其做一些简单的预处理,包括踢除孤立点、滤波去噪等操作。然后利用这些经过处理后的人体三维点云数据用Delaunay三角剖分的方法重构出三角网格化的数字化3D人体模型。然而,由于3D激光扫描仪的精度很高,使得获取的人体三维点云数据非常大,不易直接对其进行处理和重构,因此就需要对这些人体三维点云数据进行简化处理。考虑到在利用3D激光扫描仪对人体进行扫描测量的时候,采集人体表面的特征点非常密集,就会导致得到的这些人体三维点云数据之间或多或少的包含一些重复的特征信息,即这些点云数据之间会有一定的相关性。而主元分析(Principal Component Analysis,PCA)正好是一种处理这类问题的统计分析方法。它能够从多个特征指标变量的变化中找出其主要的变化方向,用较少的互不相关的综合特征指标变量来代替原始相关的特征指标变量进行分析和理解,从而使问题变得更加简单和容易处理。因此,本文随后又研究了主元分析算法,并利用该方法对获取的3D人体模型数据进行了简化,为未来无线体域网的研究做出了一点贡献。
[Abstract]:In recent years, with the development of sensor technology and wireless communication technology, taking the human body as the center, Wireless Body Area Network (WBAN), which is composed of sensor nodes used to monitor human physiological index signals, has become the focus of attention. As a new wireless communication network, Wireless body area Network (WLAN) is a new wireless communication network. It has different characteristics and requirements from other wireless communication networks. Its miniaturization and practicability make it useful in medical diagnostics. The research on wireless body area network is still in its infancy, and there is still a long way to go before the technology is mature and widely used, and the challenges are still many. There are still many key technical problems to be further studied and solved. Wireless body area network is a wireless communication network built on the human body, and the human body is the main carrier of sensor nodes to build the network. It is an indispensable part of the research work of wireless body-area network. Due to the lack of safety, high cost and low efficiency in the corresponding research experiment using real physical human body model, This requires something that can take the place of the real physical human body model in the wireless body-area network-digital 3D human body model, also known as a computer human. It is a 3D model, which is virtual by computer technology and image processing technology, which can reflect the 3D features of human body. The 3D human body model can be stored and visualized in the computer, and can be adjusted and controlled arbitrarily. Compared with the real physical body phantom, it is more secure, convenient and flexible. Therefore, it can be used to carry out all kinds of experimental simulation of wireless body area network, which greatly facilitates the research of wireless body area network. To some extent, it promotes the development of wireless body-area network. Firstly, the 3D laser scanner is used to scan and measure the human body to obtain the 3D point cloud data of the human body, and then to do some simple preprocessing, including kicking out the isolated point. Filtering and denoising are used to reconstruct the triangulated digital 3D human body model by using the Delaunay triangulation method. However, because of the high precision of 3D laser scanner, the 3D laser scanner can be used to reconstruct the 3D human body model by using the Delaunay triangulation method. Make the human body 3D point cloud data very large, it is not easy to directly process and reconstruct it, So we need to simplify these 3D point cloud data. Considering that when we use 3D laser scanner to scan and measure the human body, the feature points on the human surface are very dense. The resulting 3D point cloud data of the human body contains more or less duplicated feature information. The principal component analysis (PCA) is a kind of statistical analysis method to deal with this kind of problem. It can find out the main direction of change from the change of multiple characteristic index variables. In order to make the problem easier and easier to deal with, the problem is analyzed and understood by using fewer independent synthetic feature index variables instead of the original related feature index variables. Therefore, this paper then studies the principal component analysis algorithm. The method is used to simplify the 3D human body model data and make a contribution to the future research of wireless body area network.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN92

【共引文献】