摘要
针对室内结构化环境下的导航任务,提出一种以踢脚线为参考直线的基于单目视觉的导航算法。利用单目摄像头提取图像,运用Canny边缘检测方法和霍夫变换相结合的边缘检测算法提取结构化环境中的踢脚线,根据平行直线在图像平面的成像原理,提取出参考直线和偏航角作为导航信息,实时控制小车的运动。试验结果显示,基于该方法设计的小车可以实现安全行驶,该方法获取的导航参数受光照的影响不大,稳定可靠,能满足安全性和实时性的要求。
To achieve vision navigation task in structured indoor environments,a monocular vision algorithm is proposes based on skirting lines.By using Canny edge detection algorithm and Hough transform to detect skirting lines in structured indoor environments,and using parallel lines＇ forming principle on image plane,analyzing the lines＇ parameters to locate the mobile robot gathering reference data in lines and yaw angle,the robot control in time is achieved.Experiments show that this algorithm can realize robot＇s safety traffic and less impact by light,it could plan a safe and real time local path to navigate the robot.
出处
《西安邮电学院学报》
2012年第1期75-78,共4页
Journal of Xi'an Institute of Posts and Telecommunications
关键词
视觉导航
单目视觉
霍夫变换
边缘提取
vision navigation
monocular camera
Hough transform
edge detect
