【作者】 田正冲；

【导师】 乌日图；

【作者基本信息】 内蒙古工业大学 ， 机械设计及理论， 2007， 硕士

【摘要】 随着电子计算机科学、图像处理、图像理解理论的迅速发展,机器人视觉的实际研究与应用日益得到重视。己经不再像早期局限在工业制造自动化方面,而是广泛地应用在军事、民用、科学研究等各个领域。它们的工作环境也从单纯的室内环境变成了今天的各种环境,包括天空、地下、水下、地面等。其中地面作为人类大部分活动的场所,地面移动机器人从很早开始就得到了广泛的研究。本文以移动小车为载体,重点研究并设计了一套移动机器人双目视觉系统。该实验系统主要包括两部分:(1)视觉系统的摄像机标定;(2)视觉系统的软件功能实现。论文首先在绪论中对机器人视觉做了简要的介绍,介绍了视觉系统的硬件并描述了机器人视觉系统中的关键技术:摄像机标定和图像处理技术;其次在第二章中简单的介绍图像采集卡的原理及应用。第三章中详细的介绍了本实验系统涉及的主要图像处理技术并给出了对所采集的相关图像进行处理的实际效果;在第四章中,针对视觉系统的摄像机标定,介绍了几种常用的标定方法。在此基础上,提出一种新的简单、快速的标定方法,并介绍了本实验系统定位所需目标点的求取算法;接着介绍了移动机器人的双目视觉系统设计和控制系统。视觉系统包括摄像机标定、图像采集及数字化、图像预处理、边缘检测、目标点匹配等功能。并利用实物,对其各功能一一演示,而且详细的介绍了利用程序对摄像机标定的部分功能及步骤,并对控制系统的转弯过程进行了分析;最后在总结部分对前面所做的工作进行总结并提出了今后的研究方向。本文将理论研究成果进行了实验研究,获得满意的实验结果,验证了本文提出的利用双目视觉实现移动机器人目标物检测和避碰算法的有效性。更多还原

【Abstract】 With the rapid development of computer science, image process and recognition, people are paying more and more attention on the robot vision’s research and application. It is now not localized in commercial manufacture automation, which can be widely used in both military and civilian applications. Its working environment also changes from simple indoor circumstance to all kinds of circumstance today, including sky, underground, underwater, floor and so on. The floor as a location which most of human beings activities take place in. Floor-moving robots have been researched for a long time. In this paper ,we study and design a set of robot bi-eye vision system based on moving-car .This system include two parts: (1) camera calibrating of vision system;(2) function realization by programming the software of vision system.In the first chapter, we introduce briefly the robot vision and hardware of vision system and describe the key technologies of the robot vision system: camera calibrating and image processing technology. In the second chapter, we introduce briefly the principle and application of image collection card. In the third chapter, we introduce detailedly chief image processing technology and practical effects of some collected images concerned by the experiment system. In the fourth chapter, in allusion to camera calibrating of vision system, we introduce some kinds of favorite ways of calibration. On the basis of these methods, an easy method for camera calibration and correction are studied. And also introduce calculating method of aiming point. In the fifth chapter, we introduce the design of mobile robot’s two-eye vision system and control system. The vision system has the functions including camera calibrating, collecting and image digitalizing, detecting edge, and matching dot. In this chapter each function is demonstrated and we introduce detailedly some functions of camera calibrating by means of program. We also have analyzed the process of turn. In the last chapter, we sum up the previous work and view the future research work.In the paper, we has experimented on the outcome of study, and get the satisfied experiment result, and proved the efficiency of detecting aim and avoiding obstacle through two cameras.更多还原