基于机器视觉的宽带通讯测试板定位系统研究

发布时间：2019-07-05 06:30

【摘要】：随着电子科技的迅猛发展,尤其是正处在一个新的变革时期的宽带通讯领域,信号频率越来越高,这对集成度高、布线密集的通讯印刷电路板的检测提出了更高要求。但是目前,绝大多数的宽带通讯印刷电路板的测试工作都是通过人工在放大镜下以肉眼去寻找测试点然后用探针去接触的方式完成的。这种传统的测试方式不仅效率低,人工成本高,而且长时间注视着放大镜会让人的眼睛产生酸痛和眩晕,更严重的甚至会对视力带来不可恢复的损伤。针对这些问题,本文采用机器视觉技术,实现对宽带通讯测试板测点的定位。在这过程中,针对以下几点展开研究工作:1.以得到更为清晰的数字图像为目的,根据相机的成像原理,对视觉系统中相机的安装方式、照明方案和图像采集及其显示进行研究和分析,确定了整个视觉定位系统的结构形式和工作流程,并选用以Zynq-7000系列芯片为核心的ZedBoard开发板为硬件基础,并搭建了视觉定位平台,为提取图像特征提供必要条件。2.以实现视觉系统的嵌入式软件开发设计为目的,在ZedBoard开发板的主核移植Linux操作系统,从核设置为裸跑,并通过SDK开发工具完成了主核(CPU0)和从核(CPU1)的程序开发,以及利用MiniGUI设计开发了人机交互界面,实现了视觉定位系统的整个软件的开发。3.为精确获得宽带通讯测试板检测点位置坐标,利用测试板上四个角的Mark点,以及检测点对Mark点的相对位置关系,实现了定位。在图像处理过程中,采用粗模板+精模板的方法以及旋转卡壳算法对Mark点的特征值提取,大大减小了计算量。4.通过宽带通讯测试板定位系统的实验结果,对整个系统的重复定位进行分析和评价,验证视觉定位系统的准确性和稳定性均能满足课题中对定位系统的要求。
文内图片：

图片说明：Eye-in-Hand系统
[Abstract]:With the rapid development of electronic technology, especially in the field of broadband communication in a new period of change, the signal frequency is getting higher and higher, which puts forward higher requirements for the detection of communication printed circuit board with high integration and dense wiring. However, at present, most of the testing work of broadband communication printed circuit board is done by manually looking for the test point with the naked eye under the magnifying glass and then using the probe to contact it. This traditional test method is not only inefficient, high labor cost, but also looking at the magnifying glass for a long time will cause pain and dizziness in the eyes, and more seriously, it will even cause irreparable damage to eyesight. In order to solve these problems, machine vision technology is used to locate the measuring points of broadband communication test board. In this process, the following research work is carried out: 1. In order to get a clearer digital image, according to the imaging principle of the camera, the installation mode, lighting scheme, image acquisition and display of the camera in the visual system are studied and analyzed, the structure form and work flow of the whole visual positioning system are determined, and the ZedBoard development board based on Zynq-7000 series chip is selected as the hardware foundation, and the visual positioning platform is built. It provides the necessary conditions for extracting image features. 2. In order to realize the embedded software development and design of visual system, the Linux operating system is transferred to the main core of ZedBoard development board, the slave core is set as naked, and the program development of master core (CPU0) and slave core (CPU1) is completed by SDK development tool, and the human-computer interaction interface is designed and developed by using MiniGUI, and the whole software development of visual positioning system is realized. In order to obtain the position coordinates of the detection point of the broadband communication test board accurately, the positioning is realized by using the Mark points of the four corners on the test board and the relative position relationship between the detection points and the Mark points. In the process of image processing, the method of rough template precision template and rotating shell algorithm are used to extract the eigenvalues of Mark points, which greatly reduces the amount of computation. 4. Through the experimental results of the broadband communication test board positioning system, the repeated positioning of the whole system is analyzed and evaluated, and it is verified that the accuracy and stability of the visual positioning system can meet the requirements of the positioning system in the subject.
【学位授予单位】：武汉工程大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN41;TP391.41

【相似文献】