| Autonomous Mobile Robot (AMR)have many valuable attributes that can benefit human beings in the fields of modern life.Road detection technique is a very important part of vision navigation, which is the key technique of intelligent vehicle guidance. This thesis analyses and researches the methods of road detection based image process.In this thesis, research background, meaning and current status of vision navigation are firstly introduced, and currently algorithm of image process is analyzed; secondly, a new algorithm of camera calibration based on polynomial is introduced to calibrate the camera that is fitted on the our mobile robot.Current algorithm of road detection is analyzed; To detect road area accurately, robustly and timely, according to the special task of the thesis , based on the other people's research, we proposed a new algorithm for lane detection and obstacle detection , which is based on the combination of edge detection and color identification. This method makes use of not only edge information but also color information, which is available through edge detection and Hough transform.technology. Additionally, we made some relative optimization on the Hough transform.To get the color information of the road region,the concept of the core region information is proposed,and the rule of road edge is defined.The rule is applied to the road edge detection and the obstacle detection.A real-time video sequence capturing and data processing system of AMR based on VFW is designed and implemented in Visual C++6.0. The study of all the algorithms are running on the platform.Finally, the experimental results of the AMR under different conditions are presented. |
PDF Full Text Request