项目名称: 基于车牌合作目标单目视觉定位的汽车防撞预警系统关键技术研究
项目编号: No.61203163
项目类型: 青年科学基金项目
立项/批准年度: 2013
项目学科: 自动化学科
项目作者: 秦丽娟
作者单位: 沈阳理工大学
项目金额: 24万元
中文摘要: 汽车防撞预警系统是智能汽车技术中的一项关键技术,对保障车辆和人员安全及减少交通事故和经济财产损失意义重大。基于单目视觉的目标车辆测距作为一项重要的汽车防撞预警技术具有定位速度快,成本低廉等特点。本项目针对现有单目视觉测距技术不能检测目标车辆姿态,不能应用于起伏路况以及不适合于夜间应用的问题,提出了一种以目标车辆车牌边框为合作目标的单目视觉定位方法,该方法继承了现有单目视觉定位方法的优点,并且可以在起伏路况下检测目标车辆在三维空间的位置和姿态;该方法较传统方法的定位精度更高,并且应用不受时段限制。本项目的主要研究内容包括:复杂背景下的车牌提取技术研究;基于车牌边框几何特征的单目视觉定位、定姿、测速技术和鲁棒性研究;基于单目合作目标视觉测量的安全评价与预警模型研究。本项目的研究内容是对汽车防撞预警技术和智能汽车技术理论的丰富与补充,也可以为无人驾驶车辆导航避障研究提供一种辅助方法。
中文关键词: 汽车防撞预警系统;车牌图像识别与定位;合作目标单目视觉定位;鲁棒性分析;预警与安全评价
英文摘要: Automobile anti-collision warning system is a key technology of intelligent automobile technology. It has great significance on the automobile security, personnel safety and reduces traffic accidents and economic loss of property. As an important automobile anti-collision warning technology, target automobile positioning based on monocular vision has characteristics of high position speed, low cost and so on. Aimed at the problems for existing monocular vision measurement technology that can not detect target vehicle attitude and can not suitable for nocturnal application, this project puts forward a kind of monocular vision positioning method with the cooperative target of automobile license plate frame. This method inherits the advantages of existing monocular vision method and can detect position and attitude of the target automobile in the three-dimensional space on a rough road condition.This method has better position accuracy than the traditional method. Furthermore, the application of this method is not subject to time limits.The main research contents of this project include: research on image extraction technology for license plate under complex background; study on monocular vision positioning,attitude determination, speed measurement method and robustness based on the geometric characteristics of lic
英文关键词: Automobile anti-collision warning system;License plate image recognization and location;Cooperative target monocular vision positioning;Robustness analysis;Early warning and safety evaluation