With the development of economy, modern management methods are needed to administer traffic on account of problems of traffic transportation being more and more severe, which begets the research on Intelligent Transportation System-ITS. Driver Assistance System is an important component of Intelligent Transportation System--ITS. Driver Assistance System has bright prospects, especially in automobile safety improvement and accident avoidance measurement. Over the past decades of years, vision-based road apperceive algorithms have been used in Driver Assistance System. So the study of vision-based road apperception algorithms is very important to Driver Assistance System.The research topic of the thesis is on-board monocular vision-based road recognition. The basic thinking of the thesis is as follows. Firstly, the lane recognition calls the lane detection module. When in the steady-going stage, the lane recognition calls the lane tracking module. If the vehicle turns frequently, or the angle of the turning is big, or the road condition has changed, the lane recognition calls the detection module again. The thesis consists of two parts. In the first part, the emendation of the fisheye image and the compute of the vanishing point are studied, which provide the base for the lane recognition. In the second part, the lane detection and the lane tracking are studied mainly. In the lane detection section, after analyzing and comparing the current road detection algorithms, the edge detection algorithm and the straight line detection algorithm are studied based on the principle of the Sobel and Hough. In lane tracking part, after summarizing the current road tracking algorithms, a tracking algorithm is designed and implemented.Experiments show that the method can get a reliable result, and it is robust and real-time. The thesis is useful for the fusion of other road detection algorithms in Driver Assistance System. |