【作者】 高美芹；

【导师】 明平顺；

【作者基本信息】 武汉理工大学 ， 载运工具运用工程， 2007， 硕士

【摘要】 智能车辆(Intelligent Vehicle，Ⅳ)是当今世界车辆工程领域的研究前沿和热点。智能车辆是集环境感知、规划决策、辅助驾驶等功能于一体的综合智能系统，是计算机视觉、人工智能、控制理论和电子技术等多个技术学科交叉的产物，代表了未来车辆的发展方向，具有十分广阔的应用前景。视觉导航智能车辆已经成为当今智能车辆的发展主流，智能车辆视觉导航系统是智能交通系统的重要组成部分，关于智能车导航关键技术的研究，对于保证车辆安全行驶甚至实现自动驾驶起着至观重要的作用。由于视觉导航具有信号探测范围宽、目标信息完整、价格相对便宜及最符合人类感知方式等优点，本文提出了利用单目视觉技术在汽车辅助驾驶及自主导航系统中实现导航的方案。目前，国内外已有很多视觉导航系统，但在系统的实时性、鲁棒性和实用性方面尚不能满足人们的要求。论文主要研究内容为：本文首先对智能车辆视觉系统相关机理进行分析，获取智能车辆视觉系统在特定道路环境下的成像特点，对视觉系统相关参数进行合理化配置，这是深入开展智能车辆视觉系统研究的重要前提，具有实际的指导意义。其次，分析了汽车辅助驾驶及自主导航视觉系统的结构，将重点放在道路检测与跟踪上。通过与使用立体视觉技术的方法做对比，阐述了采用单目视觉技术的优点和实际意义。最后，在车道检测中本文提出了一种基于车道线多特征分析的车道检测算法，可以有效的去除阴影、噪声点对车道检测的影响，并且能够很方便的应用于智能汽车视觉导航中。更多还原

【Abstract】 Intelligent vehicle (IV), which is on the top level in the research field of vehicle engineering, is the hotspot around the world nowadays. Intelligent vehicle is a comprehensive intelligent system which integrates the function of environment perception, planning and decision-making, driving assistance, etc. Intelligent vehicle, which represents the direction toward the development of future vehicle, is also a product that contains multi-technical subjects including computer vision, artificial intelligence, auto-control theory, and electronic technology. Therefore, intelligent vehicle has a highly expansive application prospect.As vision-based intelligent vehicle gradually is developed to be the mainstream of various intelligent vehicles,and the research on key techniques for intelligent vehicle navigation system does great contribution to safe driving and even.Because the vision guidance system can get wide and integrate vision, costs fewer, and can fit the perceiving way of the human, this paper applies monocular vision method into the vehicle assistant driving and the automatic guidance. At present, although there are many vision guidance systems, none of them can meet our requirement in real-time, robustness, and practice. The main contents of my dissertation are as follows:Firstly,the destination of this part is to analyze the mechanism of the vision system performance, obtain the imaging characteristics of the vision system in specific environment and configure the relevant parameters rationally, which have practical significance, and provide reasonable preconditions for further investigation.Secondly,this paper analyzes the structure of the vehicle assistant driving and the automatic guidance system, and the emphasis is road detection and tracking. Compared with reappearing three-dimensional scene by stereo vision, it demonstrates the merit and practical significance of the monocular vision method.Lastly, In the lane detecting, we propose a method which fuses the way of lane edges detection and the method of color-based segmentation. The method reduces the shadow and noise points influence on lane detection and is convenient in the applications in the vision navigation of the Automatic Highway Vehicle.更多还原